I. INTRODUCTION

Why does reliance on seigniorage as a means of financing government budget deficits differ so widely across countries and across time? The traditional explanation to this question is that the capacity of governments to collect taxes determines the extent to which governments rely on the alternative of seigniorage. The capacity to collect taxes depends on economic structure as well as the efficiency of government in collecting taxes. The structure of the economy is relevant for the capacity to collect taxes because the ease of tax collection can differ by sector, and economies with a structure that is difficult to tax lowers the detection of tax evasion and encourages the development of the shadow economy. (1) Thus, in the traditional explanation, a larger shadow economy is associated with greater reliance on seigniorage. However, given the difficulty in measuring the shadow economy, the empirical literature has attempted to control for tax evasion in the shadow economy by controlling for a narrow selection of economic sectors. Such controls have included the gross domestic product (GDP) shares of manufacturing, international trade, and mining, which are considered to be relatively easy to tax, and the GDP share of agriculture, which is considered relatively more difficult to tax. (2)

While the aspect of economic structure emphasized in the traditional explanation of seigniorage relates to tax evasion across sectors, natural resource endowments have not been considered, as they have in analyses of conventional taxation. (3) Presumably, higher natural resource rents are a revenue source that will influence the extent government relies on seigniorage and other forms of taxation. Given the importance of primary commodities in the economic structure of developing countries, it is potentially important to control for natural resource rents, especially in cross-country studies that include both developed and developing countries.

A number of researchers have noted that the traditional explanation of seigniorage cannot adequately account for the variations of seigniorage across countries and across time, and propose that political economy factors also play an important role by affecting the capacity to collect taxes through the efficiency of government tax collection. Cukierman, Edwards, and Tabellini (1992) present a model where political instability and polarization endogenously determine the efficiency of tax collection. In their model, the efficiency of tax collection is chosen by the incumbent government, and this choice affects efficiency with a lag that could impact a future government's ability to collect taxes. Increased political instability increases the chance of electoral defeat of the incumbent government, so the incumbent government chooses to have a less efficient tax collection system as a means of constraining future governments that have different preferences regarding public good allocation, which reflects political polarization. They provide evidence showing that political instability and polarization are directly related to seigniorage in a cross-section of countries. Aisen and Veiga (2008) also find evidence that political instability and polarization are important determinants of seigniorage across countries and across time using panel data.

In addition to government investment in the efficiency of tax collection, another aspect of government policy which has been considered in the literature is financial sector repression. If financial repression results in a low level of financial sector development, the incentive to operate in the shadow economy is increased because the value of foregone financial services used in the formal sector is low. (4) Because shadow economy transactions tend to be more cash-intensive, lower financial development implies that the government will be more reliant on seigniorage as a means of taxing the shadow economy. A number of studies find evidence that lower financial development is associated with greater reliance on seigniorage. (5)

Summarizing, the existing empirical literature suggests that seigniorage is determined by structural, political economy, and financial development factors. However, there are a number of potential weaknesses in the existing work. As discussed above, the controls for the shadow economy seem narrow and may not adequately capture the capacity of the government to collect taxes, as a broader measure of the shadow economy would. Furthermore, natural resource rents have not been considered in the literature and it is likely this source of revenue is important for many developing countries.

There are also potentially important issues related to the econometric framework used in the literature. Most prior research is cross-sectional, which does not sufficiently address the concerns of endogeneity, omitted variables, unobserved country-specific effects, and measurement errors. The few existent studies that use panel analysis do not thoroughly address the potential endogeneity and measurement errors, and produce less efficient estimates.

The main purpose of this article is to reexamine the main explanations for the variation of seigniorage across countries and across time with a more comprehensive set of controls for economic structure and, to the best of our knowledge, the first application of dynamic panel estimation using generalized method of moment (GMM) techniques to the seigniorage literature, which rigorously deal with the problems plaguing most previous studies. We find that the shadow economy is directly related to seigniorage, as expected, and natural resource rents are an important contributor to seigniorage. Once the shadow economy and natural resource rents are controlled for, we find little evidence that political instability and polarization affect seigniorage, while we do confirm the importance of financial development. The results are robust to a number of extensions, including substantial error in the measurement of the shadow economy.

II. DATA AND EMPIRICAL MODEL

We employ a large panel dataset from 144 countries, covering the period 1970-2009, to investigate the main determinants of seigniorage. (6) Our panel is unbalanced because some countries have limited observations in the sample, and the number of countries is much larger than the number of years. Tables A1 and A2 provide the descriptive statistics and the correlation matrix for the variables discussed in this section. Seigniorage is measured by the annual change in reserve money, taken from line 14a of International Monetary Fund International Financial Statistics (IMF-IFS). In our main regressions, we normalize seigniorage by government spending, (7) instead of gross national product or government revenues. We focus on this normalization because gross national product does not include the value of production in the shadow economy, and tax revenue data are available for a smaller number of countries than government spending.

The size of the shadow economy as a share of formal output, Shadow economy 1, is obtained from new estimates of the shadow economy presented in Elgin and Oztunali (2012). This measure is based on a calibrated dynamic general equilibrium model and has a significantly larger number of observations than other estimates available in the literature. (8) As one of our robustness checks, we also use estimates from Schneider, Buehn, and Montenegro (2010), denoted Shadow economy 2, which are based on the statistical theory of unobserved variables and are generated from the multiple indicators multiple causes (MIMIC) model. (9) As explained above, we expect that a larger shadow economy will be associated with greater reliance on seigniorage.

Natural resource rents are the difference between the value of natural resources (oil, natural gas, coal, minerals, and forests) and the cost of production. The variable Natural resources is natural resource rents as a share of GDP and is obtained from the World Bank World Development Indicators (WB-WDI). Natural resource rents are often an important source of revenue to the government, especially in developing countries, so it is natural to assume that higher natural resource rents allow governments to rely less on seigniorage. (10) However, the relationship between natural resource rents and seigniorage can depend in part on the exchange rate regime, because natural resource exports are typically denominated in foreign currency. Under managed exchange rates, a larger flow of natural resource exports requires the central bank to purchase foreign exchange with domestic currency. With less than full sterilization of such foreign exchange market intervention, reserve money and seigniorage increase. Under a flexible exchange rate regime, a larger flow of exports does not require foreign exchange intervention and the greater revenues can allow the government to decrease reliance on seigniorage. Given the large number of developing countries exporting natural resources and managing exchange rates in our sample, we expect to find a direct relationship between natural resource rents and seigniorage. (11)

The exchange rate regime is measured by the de facto coarse classification of exchange rate regime from Ilzetzki, Reinhart, and Rogoff (2010). This measure, denoted by Exchange rate regime, takes on integer values between 1 and 6, where larger values indicate greater flexibility of exchange rates. (12) We extend this index to the last 2 years of our sample when possible using the IMF Annual Report on Exchange Rate Arrangements and Exchange Rate Restrictions. (13) Apart from conditioning the effect of natural resource rents on seigniorage, greater exchange rate flexibility allows the central bank to raise seigniorage with less foreign exchange reserve loss, so we expect to find a direct relationship between the exchange rate index and seigniorage.

As discussed in the previous section, the political economy approach has emphasized the role of political polarization and political instability. Political polarization is often proxied by measures of ethnic diversity, such as the ethno linguistic fractionalization (ELF) index, which measures the probability that two randomly selected individuals from a given country belong to different ethnic groups. (14) However, Montalvo and Reynal-Querol (2005a) argue that a measure of polarization is better than a measure of social fractionalization to capture the probability of conflicts, and hence political instability. They argue that it is ambiguous to what extent a diversity index can capture the intensity of potential ethnic conflict. It is notable that there is less violence in both highly homogenous and highly heterogeneous countries, and that the relationship between ethnic diversity and potential conflict may not be monotonic. (15) Thus, a polarization index--rather than a social fractionalization index--should be a better measure of ethnic diversity appropriate to capture potential conflicts among heterogeneous ethnic groups. (16) In a similar vein, Desmet, Ortuno-Ortin, and Wacziarg (2012) compute alternative measures of diversity--measured by fractionalization and polarization--at 15 different levels of aggregation, using linguistic trees. They argue that the most relevant type of cultural diversity differs for a range of political economy outcomes. For outcomes comprising conflicts of interest, such as civil wars and redistribution, deep cleavages originating at an earlier time in history and obtained at high level of aggregation are more relevant. Conversely, for outcomes involving issues of efficiency and coordination, such as economic growth, superficial cleavages that have developed more recently and obtained at low level of aggregation are most relevant. Accordingly, we use their measure of polarization with the lowest degree of aggregation to proxy for political polarization, denoted by Ethnic polarization. (17) We also use the ELF index for robustness check, in addition to alternative diversity measures of different levels of aggregation. (18) For both measures, we expect a direct relationship with seigniorage.

Political polarization is likely to be associated with greater fractionalization of the government across political parties. Accordingly, we include measures of fractionalization in the government from World Bank Database of Political Institutions (Beck et al. 2001). Legislators frac is the probability that two random draws of members of the legislature will be of different political parties. Lawmaking power is a dummy variable which takes a value of one if the executive's party controls all houses of the legislature and zero otherwise. Government frac is the probability that two random draws of deputies from among the government parties in the parliament will be of different parties. We also include a dummy variable, Military, indicating whether the chief executive is a military officer, because the military's involvement in politics is more likely under greater political polarization.

Political instability is measured by Govt turnover, which measures the percentage of veto players dropping from the government in a given year. Veto players in presidential systems are the President and the majority political party in the legislature, while in parliamentary systems, the veto players are the Prime Minister and the three largest political parties in the Parliament. (19) Because it may take time for a change in government to impact government revenue decisions, we lag this variable one period. We also consider the number of years the chief executive's party has held the chief executive's office, denoted Years in office. Alternative measures of political instability used in previous research are also considered, each lagged one period. (20) These measures are obtained from Banks and Wilson (2015). The number of fundamental changes in the state's constitutional structure in a given year is denoted by the variable Lagged constl change. Lagged cabinet change denotes the number of times in a year the premier and/or at least 50% of the cabinet posts are changed. Lagged exec change denotes the number of times in a year that the control of executive power changes, excluding cases where the new executive is not independent of their predecessor. We expect that all of these measures are directly related to seigniorage except Years in office, which we expect to be inversely related.

The influence of political economy factors on seigniorage depends on the independence of the central bank, so we also include the index of de jure central bank autonomy, CBA, calculated by Amone et al. (2009). This index measures CBA across two dimensions: autonomy to choose the objectives of monetary policy and autonomy to choose its instruments. This index has broad coverage, particularly for developing countries. (21) We expect this index to be associated with lower seigniorage. As a robustness check, we also utilize data on the average turnover rate of central bank governors over the period 1970-2004, (Dreher, Sturm, and de Haan 2008). (22)

We consider three financial indicators that reflect the extent of financial development. Capital account openness is an indicator of less financial repression, measured by the capital account openness index presented in Chinn and Ito (2006), denoted Financial openness. A larger value of the index indicates that a country is more open to cross-border transactions. Credit to private is domestic credit to the private sector as percentage of GDP from the WB-WDI. We expect these two indicators to be inversely related to seigniorage. The variable M2 denotes money and quasi money, from lines 34 and 35 of the IMF-IFS, as a percentage of GDP. This variable is a common indicator of financial development, however because of the direct relationship between reserve money and money supply through the money multiplier, it is possible that M2 is directly related to seigniorage.

We also control for a number of common additional factors, including lagged inflation, (23) lagged real GDP growth, and government debt as a share of GDP, all obtained from WB-WDI. Lagged inflation is expected to be directly related to seigniorage because of lagged adjustment of the inflation tax base, the real money stock. (24) Lagged GDP growth affects the size of the conventional tax base and decreases the incentive to rely on seigniorage. Another reason to consider lagged output is that current output, by affecting money demand and interest rates, could induce a central bank with an interest rate operating target to change reserve money for this reason and not because of seigniorage considerations. The level of government debt can increase the incentive to increase seigniorage because the resulting inflation lowers the real value of the debt, so we expect a direct relationship between government debt as a share of GDP, denoted by Government debt, and seigniorage.

Period dummy variables are also included to remove universal time-related shocks from the errors. (25) Because global economic slowdowns can impact seigniorage and may not be captured by other explanatory variables, we have three dummy variables for the periods 1990-1991, 2001, and 2008-2009.26 To account for the possibility that the recession dummies do not adequately capture time effects, we include a quadratic time trend. As a robustness check, we also consider decade dummies.

The panel data analysis we conduct here has numerous benefits over cross-sectional analysis utilized in most prior work on the determinants of seigniorage. In particular, it takes into consideration how the shadow economy within a country over time can have an influence on the country's dependence on seigniorage, in addition to considering the cross-country relation between the shadow economy and seigniorage. Moreover, it controls for unobserved, time-invariant, country-specific effects, which reduces the estimated-coefficients biases, while those country-specific effects would be part of the error term in cross-country regressions (Levine, Loayza, and Beck 2000). Furthermore, the panel estimator can control for the potential endogeneity of all explanatory variables by employing instrumental variable (IV) techniques.

We estimate the following dynamic panel data model:

(1) [S.sub.i,t] = [PHI][RM.sub.i,t-1] + [delta][[pi].sub.i, t-1] + [lambda][SE.sub.i,t] + [X'.sub.i,t] [beta] + [[mu].sub.i] + [v.sub.i,t]

where [S.sub.i,t] is seigniorage as a share of government spending for country i in year t, [RM.sub.i,t-1] is lagged reserve money for country i, normalized by government spending, [[pi].sub.i,t-1] is lagged country inflation, [SE.sub.i,t] is the output of the shadow economy as a share of formal economy output, [X.sub.i,t] is a vector of explanatory variables, including political economy and financial variables, [[mu].sub.i] represents unobservable, time-invariant factors for each country, and vi t is the error term. The [[mu].sub.i,t] term is assumed to be random and independent of [v.sub.i,t], and that [[mu].sub.i] ~ IID (0, [[sigma].sup.2.sub.[mu]]) and [v.sub.i,t] ~ IID (0, [[sigma].sup.2.sub.v]).

The model allows for the persistence of seigniorage over time due to the presence of both the lagged dependent variable as a regressor and country effects. It captures country heterogeneity and allows for endogeneity of all explanatory variables with the country effects by employing the IV technique (Mundlak 1978; Baltagi 2013, 155-168). (27) Including the lagged dependent variable among the regressors also has the advantage of controlling for all time-invariant structural characteristics, as well as observed and unobserved historical factors, that influence its current level (Wooldridge 2011, 371-374).

The ordinary least squares (OLS) and the random effects estimators of Equation (1) are biased and inconsistent due to the autocorrelation resulting from the presence of the lagged dependent variable among explanatory variables, which is correlated with the error term. (28) The fixed effects estimator is also biased and inconsistent when the number of countries is large and number of time periods is small (Baltagi 2013, 155-156). The OLS and within estimators do not take into consideration the endogeneity of the lagged dependent variable. Arellano and Bond (1991) have suggested a GMM procedure to estimate this model, which uses the lagged values of the endogenous explanatory variables as instruments to address the endogeneity problem. Arellano and Bover (1995) and Blundell and Bond (1998) have developed a unifying GMM framework for considering efficient IV estimators. We utilize the Blundell and Bond (1998) system GMM estimation, which achieves higher estimation efficiency, to estimate Equation (1). The system GMM estimator deals with important modeling concerns, such as the endogeneity of explanatory variables and fixed effects (Roodman 2009a). (29) We begin by first-differencing each variable to eliminate country fixed effects:

(2) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

We then simultaneously estimate the equations in differences and in levels, using distinct instruments for each equation (lagged variables in levels instrument the differenced equation, and lagged differences instrument the levels equation), which adequately deals with suspected endogeneity.

III. EMPIRICAL RESULTS

The Blundell and Bond (1998) two-step system GMM estimator is utilized to estimate Equation (1), applying the robust standard errors proposed by Windmeijer (2005). We are treating both lagged reserve money and the exchange rate regime as endogenous variables in the GMM-style instruments, with lags 2 and longer for the transformed equation and lag 1 for the levels equation--the standard treatment for endogenous variables (Roodman 2009a, 2009b). To avoid the instrument proliferation problem, we restrict the lag ranges used in generating these instrument sets to 10, instead of all available lags (Roodman 2009a). (30) The collapse option was also invoked to reduce the number of moment conditions. In each specification, the diagnostics are satisfactory. The Arellano-Bond test for first-order serial correlation in the disturbances--in differences--rejects the null that there is no first-order serial correlation, but it does not reject the null of no second-order serial correlation, as shown in Tables 1 and 2. The Hansen test of over-identified restrictions for each specification does not reject the null hypothesis that all instruments are valid, and the instrument count is considerably less than the number of countries. The difference-in-Hansen test of exogeneity of the full set of GMM instruments for the levels equation does not reject the null that instruments are exogenous. (31)

Table 1 presents the results of the system GMM estimator described above, using the moment conditions on equations in levels, in addition to the moment conditions on the first-differenced equation. The dependent variable is Seigniorage 1, measured as the change in reserve money as a percentage of government spending. The first column reports a base model that includes the level of lagged reserve money, measured as percentage of government spending (Lagged RM 1), Lagged inflation, Lagged GDP growth, Exchange rate regime, Government debt, Natural resources, Ethnic polarization, Shadow economy 1, the period dummy variables, and a quadratic time trend. Columns (2) through (4) add political instability variables to the base model. Column (2) adds Lagged govt turnover and Military to the base model. Columns (3) and (4) consider constitutional changes and two measures of turnover in government, Lagged cabinet change and Lagged exec change. Column (5) includes the measure for central bank independence and columns (6) and (7) include the M2 measure of financial development with different sets of political instability variables.

The results are generally consistent. Lagged reserve money coefficient estimates are positive and statistically significant in each specification, indicating a dynamic panel. (32) The coefficient on the exchange rate regime is positive and significant in all regressions, indicating that the fixed exchange rate regime constrains the government from relying on seigniorage to finance its deficits. The coefficients on the economic structure variables Natural resources and Shadow economy 1 are consistently positive and significant. The direct relationship between natural resource rents to the government and seigniorage is consistent with the fact that such rents are denominated in foreign currency and many of the countries in our sample manage exchange rates. (33) An increase in natural resource rents requires the central bank to purchase foreign exchange earnings from natural resources, increasing reserve money. The coefficient estimates are of similar magnitude as those related to the shadow economy and suggest that this factor is an important feature of economic structure which has not been considered in the previous literature. As expected, a larger shadow economy implies that the taxable capacity of the government is lower, so there is a greater reliance on seigniorage. The M2 measure of financial development is positive and significant at the 10% level. As stated above, this common measure of financial development can be positively related to seigniorage because the money supply is related to reserve money through the money multiplier. We further investigate the role of financial development below.

It is notable that no indicators of political instability or polarization are statistically significant, in contrast to previous findings, and generally have negative coefficients, contrary to expectations. While our dynamic analysis finds no evidence that variations in political polarization and instability increase seigniorage, we do not interpret this result as necessarily implying that such political economy factors are irrelevant. We do not rule out the possibility that political economy factors are responsible for the initial decreased tax capacity, development of the shadow economy, and increased reliance on seigniorage. However, if there are barriers to reversing the size of the shadow economy, which seems likely, then a decrease in political polarization and instability after the shadow economy has developed might not affect seigniorage. This interpretation seems to be supported by cross-section models that regress long-term average seigniorage measures on long-term averages of political economy variables. (34) In none of the regressions is the coefficient on CBA significant. Perhaps, this is not surprising given that we find no support for political polarization and instability being determinants of seigniorage. (35)

Table 2 reports the estimation results when we control for additional measures of financial development, Financial openness, and Credit to private. In general, the main results from Table 1 continue to hold: Seigniorage increases with natural resource rents and the size of the shadow economy, while there is little support for variations in political instability and polarization being important in explaining variations in seigniorage. All of the political instability and polarization variables are insignificant, except Lagged constl change, which is significant at the 10% level in two cases, but with a negative sign. The coefficient on Financial openness is consistently negative but insignificant, while Credit to private is consistently negative and significant, providing mixed support for the idea that increased financial development facilitates the avoidance of seigniorage.

Table 3 extends the analysis in two ways. First, we extend our analysis of political polarization to measures of fractionalization of the government across political parties. We also consider an alternative measure of turnover of the executive, Years in office. Columns (1) and (2) show that the inclusion of this alternative measure of executive turnover leads to similar results as above, with none of the political polarization and instability variables significant. Columns (3) to (8) include the fractionalization variables Lawmaking power, Legislators frac, and Government frac. The coefficients on all of these variables are negative and none are significant.

IV. SENSITIVITY ANALYSIS

A main concern in using measures of the shadow economy in an empirical study of seigniorage is the difficulty of obtaining accurate estimates of the shadow economy. We test the sensitivity of our results to measurement error in the estimates of the shadow economy by adding randomly generated values of [+ or -] 10% to Shadow economy 1 and re-running the models of Tables 1 and 2. The results are reported in Tables 4 and 5. The main results continue to hold: The shadow economy and natural resource rents have an important direct effect on seigniorage, while we find no evidence to support political economy influence and evidence in favor of the hypothesis that greater financial development decreases seigniorage. The results are consistent with those in Tables 1 and 2, implying that a reasonable measurement error in the shadow economy estimates does not drive the results. (36)

We also consider an alternative measure of seigniorage, the change in monetary base as percentage of nominal GDP, Seigniorage 2. While we have reservations about the usefulness of this measure, given the wide variations in the size of the shadow economy across countries, this measure has been used in a number of studies and we present these regressions to facilitate comparison. The results are presented in Tables 6 and 7. The results are generally consistent with the results in Tables 1 and 2, but with some notable differences. In both tables, we see that Lagged constl change is significant in all regressions, but again with a negative sign. The other main difference in the results reported in Tables 6 and 7 is that none of the financial development variables are significant, in contrast to the results reported in Tables 1 and 2. The Arellano-Bond tests for second-order serial correlation in the idiosyncratic disturbance term [v.sub.i,t], applied to residuals in differences, are satisfied in each specification, as are the Hansen test of over-identifying restrictions and the difference-in-Hansen test.

Next, we investigate if the results are robust to another measure of the shadow economy that has been used in the literature. Shadow economy 2 is a measure of the shadow economy estimated from a MIMIC model. The estimates come from Schneider, Buehn, and Montenegro (2010) and are available from 1999 to 2007. We estimate the models of Table 1 using this alternative measure of the shadow economy and report the results in Table 8. The main results continue to hold: Natural resource rents and the shadow economy are directly related to seigniorage, and there is little support for political economy factors being important for the determination of seigniorage. All political instability and polarization variables are insignificant except Ethnic polarization, which is positive and significant in columns (1) through (4) at the 10% level and in column (5) at the 5% level. We again find mixed support for the idea that financial development negatively affects seigniorage. In columns (6) and (7) M2 has a negative coefficient and is insignificant, and if we estimate the models of Table 2 with the alternative measure of the shadow economy, we find that Financial openness and Credit to private have negative coefficients that are significant at the 10% level in four of the nine models. (37) These conclusions are unchanged if we introduce noise into this measure of the shadow economy, as shown in Table 9. In Table 10, we show the results using Shadow economy 2 and the alternative dependent variable Seigniorage 2. Again, natural resource rents and the shadow economy are directly related to seigniorage, and none of the political instability or polarization variables are significant. CBA is no longer insignificant, but is statistically significant only at the 10% level and the sign is positive, contrary to expectations.

In all specifications above, we have used recession dummies and a quadratic time trend to capture time effects. Given the decreased reliance on seigniorage in recent decades, we follow Aisen and Veiga (2008) and include decade dummies in the models of Table 1. The results are reported in Table 11. The decade variables for the 1980s and 1990s have negative coefficients and are generally significant, reflecting the decrease in governments' reliance on seigniorage. The remaining results are very similar to those reported in Table 1. (38) We also investigated whether our results concerning the political determinants of seigniorage are driven by the fact that our sample includes data from the low seigniorage 2000s, in contrast to other studies. (39) Table 12 reports the estimates of the models in Table 1 for the period 1970-1999. (40) Ethnic polarization and Lagged constl change have negative coefficients and are significant in all cases and Military has a negative coefficient and is significant at the 10% level in two cases. As measures of political instability and polarization, these results are contrary to the theory that greater political instability and polarization increase seigniorage. All other political variables are insignificant. We conclude that for this time period greater political instability and polarization is not associated with greater reliance on seigniorage, in line with our earlier results. (41)

We have also performed another robustness test to make sure that outliers associated with high inflation are not influencing our estimation results, where we have removed from the dataset all observations of periods for which inflation, measured by using either GDP deflator or consumer price index, was equal to or above 100% in a given year. (42) The results of the re-estimation of the main models are in the line with those obtained in Tables 1 and 2. (43)

Moreover, we test our results of the system GMM for robustness to reductions in the instrument set by cutting the instrument count and examine the behavior of coefficient estimates, in addition to Hansen and difference-in-Hansen tests, as suggested by Roodman (2009a, 2009b). Limiting the maximum number of lags used as instruments to four (instead of ten) decreases the number of instruments significantly and avoids instrument proliferation, giving consistent results. We have also treated only the lagged dependent variable (lagged reserve money) as endogenous, using its second and deeper lagged values (limited to ten to avoid proliferation of instruments) as GMM-style instruments for the difference equation, and the first and deeper lagged differenced values (capped at ten) as instruments for the level equation and the results are similar to those of Tables 1 and 2. All regressions satisfy the specification tests; there is no evidence of second-order serial correlation and all regressions pass the Hansen and the difference-in-Hansen tests. Our main results are reasonably robust to variations in the instrument set and lags used. As there is no consensus on the number of lags to use as instruments, it is also of interest to compare our system GMM results with fixed effects results to determine if the pattern of statistical significance is similar, even if the latter are known to be biased and inconsistent. We report the fixed effects estimates of the models of Tables B1 and B2 in Appendix B. The results are similar.

For further robustness tests, we control for additional variables such as country credit worthiness, (44) country terms of trade, (45) political stability and absence of violence (46) and military expenditures (47) but each is statistically insignificant. (48) We also test the robustness of the results to alternative variable definitions and measures to ensure that variable definitions do not affect the results. We employ the de jure, rather than the de facto, coarse classification of the exchange rate regime of the IMF, an alternative measure of government debt (as clarified earlier), the ELF index (instead of the ethnic polarization index), various diversity measures at different levels of aggregation, and alternative measures of religious polarization and religious fractionalization. Including these variables does not affect our results.

V. CONCLUSION AND POLICY IMPLICATIONS

The main innovations of the article are that we more comprehensively control for economic structure and use the first application of GMM estimation in a dynamic panel model in this literature, rigorously addressing the likely endogeneity and measurement errors, in addition to other problems plaguing much of prior studies. In contrast with the literature, we control for economic structure by using a model-based measure of the shadow economy, instead of more narrow measures. We also control for natural resource rents, which vary widely across countries. We find that a larger shadow economy is associated with greater reliance on seigniorage, consistent with traditional explanations for seigniorage. However, natural resource rents are directly related to seigniorage. With these more comprehensive controls and more advanced estimation techniques, we re-examine determinants of seigniorage that have been previously considered in the literature. While we find evidence confirming that greater financial development is inversely related to seigniorage, we generally find little evidence that political instability and polarization lead to greater reliance on seigniorage. We conduct a number of robustness checks and find that our results are robust to a number of alternative specifications, most notably allowing for substantial imprecision in the measurement of the shadow economy.

The results have implications for the question of how to reduce governments' reliance on seigniorage. Reforms lowering the incentives to operate in the shadow economy, including financial market liberalization, are an effective method of lowering seigniorage. Of course, such reforms are political decisions, and while our dynamic analysis suggests no immediate effect of the political environment on seigniorage, as some previous work does, (49) this does not imply that improvement in the political environment would not reduce seigniorage, as suggested by previous cross-sectional studies. (50) Indeed, a sustained decrease in political instability and polarization is likely to improve the efficiency of tax collection, reduce financial repression, decrease the size of the shadow economy, and lower reliance on seigniorage. A more immediate policy implication concerns the selection of the exchange rate regime. While exchange rate management is often seen as a device to discipline a government from relying on seigniorage, our results suggest that exchange rate management may be counterproductive for efforts to reduce seigniorage in countries with substantial natural resource rents.

ABBREVIATIONS

CBA: Central Bank Autonomy

ELF: Ethno Linguistic Fractionalization

GDP: Gross Domestic Product

GMM: Generalized Method of Moment

IMF-IFS: International Monetary Fund International Financial Statistics

IV: Instrumental Variable

MIMIC: Multiple Indicators Multiple Causes

NBER: National Bureau of Economic Research

OECD: Organization for Economic Co-operation and Development

OLS: Ordinary Least Squares

WB-WDI: World Bank World Development Indicators

doi: 10.1111/ecin.12298

Online Early publication November 25, 2015

APPENDIX A

TABLE A1
Descriptive Statistics of the Variables Included in
the Models

                                         Standard
Variable                Obs.     Mean    Deviation   Minimum   Maximum

Seigniorage 1           4,427   0.117      0.165     -0.461     0.896
Seigniorage 2           4,958   0.018      0.028     -0.050     0.286
RM 1 (-1)               4,400   0.837      0.581      0.003     3.956
RM 2 (-1)               4,921   0.122      0.088      0.001     0.805
Shadow economy 1        5,566   0.345      0.137      0.081     0.803
Shadow economy 2        1,420   0.331      0.128      0.081     0.683
Inflation (-1)          6,239   0.242      1.178     -0.319    22.517
GDP growth (-1)         6,293   0.038      0.052     -0.198     0.297
Exchange rate regime    5,775   2.160      1.331        1         6
Government debt         5,650   0.093      0.199     -0.747     1.998
Natural resources       6,630   0.082      0.147        0       1.155
Ethnic polarization     7,878   0.445      0.269        0       0.988
Govt turnover (-1)      5,120   0.121      0.287        0         1
Military                5,570   0.209      0.406        0         1
Constl change (-1)      6,401   0.082      0.287        0         4
Cabinet change (-1)     6,396   0.428      0.579        0         5
Exec change(-1)         6,407   0.183      0.444        0         4
Years in office         4,387   11.287    12.294        1        100
Lawmaking power         4,639   0.605      0.489        0         1
Legislators frac        4,654   0.469      0.303        0         1
Government frac         4,713   0.188      0.271        0         1
CBA                     6,123   0.585      0.195      0.19        1
Financial openness      5,752   -0.022     1.522     -1.864     2.439
Credit to private       5,284   0.316      0.235      0.006     0.998
M2                      5,046   0.393      0.216      0.008       1

TABLE A2
The Correlation Matrix for the Variables of the Study

                            (1)     (2)     (3)     (4)      (5)

(1) Seigniorage 1           1
(2) Seigniorage 2           0.88    1
(3) RM 1 (-1)               0.46    0.23    1
(4) RM 2 (-1)               0.23    0.24    0.72    1
(5) Shadow economy 1        0.13    0.11    0.09   -0.07     1
(6) Shadow economy 2        0.12    0.11    0.07   -0.07     0.99
(7) Inflation (-1)          0.13    0.13   -0.03   -0.05     0.05
(8) GDP growth (-1)         0.03   -0.02    0.08   -0.003    0.04
(9) Exchange rate regime    0.10    0.06    0.05   -0.06    -0.06
(10) Government debt        0.15    0.14    0.21    0.21     0.06
(11) Natural resources      0.11    0.12    0.02   -0.001    0.05
(12) Ethnic polarization    0.04    0.07    0.04    0.08     0.12
(13) Govt turnover (-1)    -0.02    0.01    0.03    0.09    -0.06
(14) Military               0.09    0.07    0.09    0.07    -0.03
(15) Constl change (-1)     0.07    0.07    0.08    0.06     0.07
(16) Cabinet change (-1)   -0.01    0.01    0.01    0.03     0.01
(17) Exec change (-1)      -0.01    0.005  -0.04   -0.02    -0.07
(18) Years in office        0.09    0.05    0.12    0.03     0.002
(19) Lawmaking power        0.12    0.06    0.24    0.18     0.14
(20) Legislators frac      -0.11   -0.04   -0.17   -0.08    -0.09
(21) Government frac       -0.06   -0.05   -0.10   -0.02    -0.22
(22) CBA                   -0.04    0.04   -0.13   -0.004   -0.18
(23) Financial openness    -0.11   -0.10   -0.01    0.05    -0.04
(24) Credit to private     -0.21   -0.17   -0.18   -0.03    -0.36
(25) M2                    -0.04    0.001   0.15    0.38    -0.39

                            (6)      (7)     (8)     (9)     (10)

(1) Seigniorage 1
(2) Seigniorage 2
(3) RM 1 (-1)
(4) RM 2 (-1)
(5) Shadow economy 1
(6) Shadow economy 2        1
(7) Inflation (-1)          0.07     1
(8) GDP growth (-1)        -0.01    -0.10    1
(9) Exchange rate regime   -0.07     0.19   -0.16    1
(10) Government debt        0.06     0.05   -0.13    0.33    1
(11) Natural resources      0.05     0.29    0.19   -0.08   -0.17
(12) Ethnic polarization    0.11     0.02    0.09   -0.13   -0.13
(13) Govt turnover (-1)    -0.04    -0.05   -0.07    0.003   0.07
(14) Military              -0.04     0.03    0.02   -0.01   -0.03
(15) Constl change (-1)     0.07    -0.02    0.05    0.06    0.09
(16) Cabinet change (-1)    0.02     0.01   -0.09    0.06    0.03
(17) Exec change (-1)      -0.06    -0.06   -0.12    0.02    0.01
(18) Years in office       -0.003    0.10    0.04    0.04   -0.16
(19) Lawmaking power        0.11     0.14    0.15   -0.12   -0.09
(20) Legislators frac      -0.06    -0.05   -0.06    0.05    0.10
(21) Government frac       -0.21    -0.10   -0.06    0.09    0.05
(22) CBA                   -0.17    -0.11   -0.03    0.09   -0.02
(23) Financial openness    -0.03    -0.17   -0.02    0.05    0.02
(24) Credit to private     -0.35    -0.30   -0.05   -0.05   -0.09
(25) M2                    -0.38    -0.25   -0.09   -0.03    0.11

                            (11)    (12)    (13)    (14)    (15)

(1) Seigniorage 1
(2) Seigniorage 2
(3) RM 1 (-1)
(4) RM 2 (-1)
(5) Shadow economy 1
(6) Shadow economy 2
(7) Inflation (-1)
(8) GDP growth (-1)
(9) Exchange rate regime
(10) Government debt
(11) Natural resources      1
(12) Ethnic polarization    0.12    1
(13) Govt turnover (-1)    -0.03    0.01    1
(14) Military               0.17   -0.04   -0.11    1
(15) Constl change (-1)     0.03    0.04    0.03    0.13    1
(16) Cabinet change (-1)   -0.04    0.08    0.03   -0.02    0.07
(17) Exec change (-1)      -0.06    0.02    0.08   -0.10    0.10
(18) Years in office        0.13    0.02   -0.13    0.18   -0.001
(19) Lawmaking power        0.19    0.10   -0.08    0.22    0.09
(20) Legislators frac      -0.03   -0.04    0.08   -0.24   -0.01
(21) Government frac       -0.11   -0.01    0.04   -0.07   -0.03
(22) CBA                   -0.11   -0.11    0.07   -0.06    0.03
(23) Financial openness    -0.06   -0.04    0.01   -0.01    0.02
(24) Credit to private     -0.24   -0.06    0.04   -0.13   -0.09
(25) M2                    -0.23   -0.05    0.07   -0.10   -0.06

                            (16)    (17)    (18)    (19)    (20)

(1) Seigniorage 1
(2) Seigniorage 2
(3) RM 1 (-1)
(4) RM 2 (-1)
(5) Shadow economy 1
(6) Shadow economy 2
(7) Inflation (-1)
(8) GDP growth (-1)
(9) Exchange rate regime
(10) Government debt
(11) Natural resources
(12) Ethnic polarization
(13) Govt turnover (-1)
(14) Military
(15) Constl change (-1)
(16) Cabinet change (-1)    1
(17) Exec change (-1)       0.45    1
(18) Years in office        -0.10   -0.13   1
(19) Lawmaking power        -0.12   -0.18   0.32    1
(20) Legislators frac       0.15    0.16    -0.37   -0.60   1
(21) Government frac        0.06    0.09    -0.31   -0.48   0.61
(22) CBA                    0.11    0.16    -0.28   -0.30   0.42
(23) Financial openness     0.04    0.08    -0.03   -0.19   0.18
(24) Credit to private      -0.01   0.04    -0.14   -0.25   0.15
(25) M2                     -0.01   0.03    -0.11   -0.11   0.10

                            (21)   (22)    (23)    (24)    (25)

(1) Seigniorage 1
(2) Seigniorage 2
(3) RM 1 (-1)
(4) RM 2 (-1)
(5) Shadow economy 1
(6) Shadow economy 2
(7) Inflation (-1)
(8) GDP growth (-1)
(9) Exchange rate regime
(10) Government debt
(11) Natural resources
(12) Ethnic polarization
(13) Govt turnover (-1)
(14) Military
(15) Constl change (-1)
(16) Cabinet change (-1)
(17) Exec change (-1)
(18) Years in office
(19) Lawmaking power
(20) Legislators frac
(21) Government frac        1
(22) CBA                    0.34   1
(23) Financial openness     0.10   0.26    1
(24) Credit to private      0.23   0.06    0.30    1
(25) M2                     0.26   -0.04   0.21    0.76    1

TABLE A3
Descriptive Statistics of Shadow Economy and Natural
Resource Rents for Various Country-Income Groups

Country-Income                       Standard
Group                Obs.    Mean    Deviation   Minimum   Maximum

Shadow economy 1
  OECD               1,158   0.198     0.078      0.081     0.620
  Hi-income          1,762   0.224     0.096      0.081     0.745
  Non-OECD           4,538   0.385     0.123      0.107     0.803
  Non-hi-income      3,947   0.402     0.116      0.115     0.803
  Low and lower-     2,526   0.424     0.112      0.139     0.803
    middle-income
  LDC                1,415   0.435     0.093      0.230     0.784

Shadow economy 2
  OECD                278    0.181     0.061      0.081     0.327
  Hi-income           450    0.213     0.088      0.081     0.540
  Non-OECD           1.142   0.367     0.113      0.111     0.683
  Non-hi-income       979    0.385     0.105      0.119     0.683
  Low and lower-      620    0.404     0.100      0.144     0.683
    middle-income
  LDC                 336    0.409     0.069      0.266     0.586

Natural resources
    rents
  OECD               1,199   0.018     0.033        0       0.221
  Hi-income          2,424   0.078     0.170        0       1.155
  Non-OECD           5,926   0.095     0.157        0       1.155
  Non-hi-income      4,736   0.086     0.136        0       1.151
  Low and lower-     2,878   0.083     0.117        0       1.059
    middle-income
  LDC                1,619   0.079     0.114        0       1.031

Notes: Country classification is based on the 2012 main
World Bank country classification, available at http://
data.worldbank .org/about/country-and-lending-groups. It is
based on gross national income (GNI) per capita. Low/income
economies are defined as those with a GNI per capita of
$1,035 or less in 2012; lower/middle/income economies: more
than $1,035 but less than $4,085; and hi/income economies:
$12,616 or more. The least developed countries (LDC) is a
widely used United Nations classification, which includes 48
countries, as of 2012 (for complete list and criteria of
identification, see http://www/un/org/en/development/desa/
policy/cdp/ldc/ldc_criteria.shtml).

APPENDIX B

TABLE B1
The Fixed Effects Estimation Results of Table 1 Models

                                 (1)          (2)          (3)

Lagged RM 1 (-1)              0.060 ***    0.054 ***    0.060 ***
                               (0.007)      (0.008)      (0.008)
Lagged inflation (-1)         0.019' **    0.021 ***    0.019 ***
                               (0.003)      (0.003)      (0.003)
Lagged GDP growth (-1)          -0.001       -0.031       -0.003
                               (0.055)      (0.062)      (0.056)
Exchange rate regime          0.020 ***    0.026 ***    0.020 ***
                               (0.003)      (0.003)      (0.003)
Government debt                 0.041        0.033       0.042 **
                               (0.020)      (0.022)      (0.021)
Natural resources             0.291 ***    0.358 ***    0.282 ***
                               (0.047)      (0.053)      (0.047)
Shadow economy 1              0.222 ***      0.155      0.218 ***
                               (0.077)      (0.096)      (0.079)
Lagged govt turnover (-1)                    0.002
                                            (0.009)
Military                                     0.004
                                            (0.011)
Lagged constl change (-1)                                 0.002
                                                         (0.009)
Lagged cabinet change (-1)                                -0.006
                                                         (0.005)
Lagged exec change(-1)

M2

1990-1991 dummy                0.021 *      0.022 *      0.022 *
                               (0.011)      (0.011)      (0.011)
2001 dummy                      0.004        0.005        0.007
                               (0.013)      (0.013)      (0.013)
2008-2009 dummy                 -0.006       -0.011       -0.007
                               (0.013)      (0.014)      (0.013)
Quadratic time trend            -0.100       -0.054       -0.088
                               (0.071)      (0.080)      (0.073)
Constant                      -0.077 **    -0.072 **    -0.075 **
                               (0.031)      (0.038)      (0.032)
[R.sup.2] (within)              0.077        0.083        0.078
F-test                        25.13 ***    20.17 ***    20.69 ***
Total panel obs.                3,463        3,053        3,349

                                 (4)          (5)          (6)

Lagged RM 1 (-1)              0.060 ***    0.049 ***    0.055 ***
                               (0.008)      (0.008)      (0.008)
Lagged inflation (-1)         0.019 ***    0.022 ***    0.020 ***
                               (0.003)      (0.003)      (0.003)
Lagged GDP growth (-1)          0.005        -0.048       -0.023
                               (0.056)      (0.063)      (0.057)
Exchange rate regime          0.020' **    0.025 ***    0.019 ***
                               (0.003)      (0.003)      (0.003)
Government debt                0.041 **      0.032        0.050
                               (0.021)      (0.025)      (0.023)
Natural resources             0.281 ***    0.417 ***    0.334 ***
                               (0.047)      (0.053)      (0.047)
Shadow economy 1              0.218 ***    0.268 ***    0.335 ***
                               (0.079)      (0.103)      (0.085)
Lagged govt turnover (-1)                    0.004
                                            (0.009)
Military                                     0.009
                                            (0.011)
Lagged constl change (-1)       -0.002                    0.002
                               (0.009)                   (0.009)
Lagged cabinet change (-1)                                -0.003
                                                         (0.005)
Lagged exec change(-1)          0.004
                               (0.006)
M2                                         0.173 ***    0.167 ***
                                            (0.030)      (0.028)
1990-1991 dummy                 0.021        0.018        0.018
                               (0.011)      (0.011)      (0.011)
2001 dummy                      0.007        0.004        0.006
                               (0.013)      (0.014)      (0.014)
2008-2009 dummy                 -0.006     -0.030 **      -0.023
                               (0.013)      (0.014)      (0.014)
Quadratic time trend            -0.094      -0.207"     -0.244 ***
                               (0.073)      (0.087)      (0.079)
Constant                      -0.078 **    -0.168 ***   -0.172 ***
                               (0.032)      (0.044)      (0.038)
[R.sup.2] (within)              0.077        0.098        0.094
F-test                        20.59 ***    21.01 ***    22.12 ***
Total panel obs.                3,349        2,856        3,151

Notes: The dependent variable (Seigniorage 1) is measured as
the change in reserve money, divided by government spending.

RM 1 is reserve money (line 14a of IMF-IFS) as percentage of
government spending. Ethnic polarization and CBA are omitted
because of collinearity, and hence specification 5 of Table
1 is excluded.

* p< .1; ** p< .05; *** p< .01.

TABLE B2
The Fixed Effects Estimation Results of Table 2 Models

                                 (1)          (2)          (3)

Lagged RM 1 (-1)              0.051 ***    0.056 ***    0.047 ***
                               (0.009)      (0.008)      (0.009)
Lagged inflation (-1)         0.024 ***    0.021 ***    0.024 ***
                               (0.003)      (0.004)      (0.004)
Lagged GDP growth (-1)          -0.059       -0.038       -0.032
                               (0.064)      (0.065)      (0.066)
Exchange rate regime          0.025 ***    0.025 ***    0.026 ***
                               (0.003)      (0.004)      (0.004)
Government debt                 0.027       0.044 *       0.004
                               (0.025)      (0.024)      (0.026)
Natural resources             0.408 ***    0.373 ***    0.399 ***
                               (0.054)      (0.055)      (0.055)
Shadow economy 1              0.270 ***      0.111       0.232 **
                               (0.104)      (0.107)      (0.109)
Lagged govt turnover (-1)       0.003        0.001       -0.0002
                               (0.009)      (0.010)      (0.010)
Military                        0.014        0.006       0.019 *
                               (0.011)      (0.011)      (0.011)
Lagged constl change (-1)

Lagged cabinet change (-1)

Lagged exec change (-1)

Financial openness            -0.011 ***    -0.007 *    -0.008 **
                               (0.004)      (0.004)      (0.004)
Credit to private                            -0.024     -0.185 ***
                                            (0.027)      (0.035)
M2                            0.183 ***                  0.324" *
                               (0.031)                   (0.044)
1990-1991 dummy                 0.017       0.023 *       0.017
                               (0.011)      (0.012)      (0.012)
2001 dummy                      0.004        0.006        0.001
                               (0.014)      (0.014)      (0.014)
2008-2009 dummy               -0.029 **     -0.027 *     -0.026 *
                               (0.015)      (0.015)      (0.016)
Quadratic time trend            -0.140       -0.019     -0.197 **
                               (0.091)      (0.093)      (0.096)
Constant                      -0.178 ***     -0.051     -0.156 ***
                               (0.045)      (0.046)      (0.048)
[R.sup.2] (within)              0.104        0.090        0.111
F-test                        20.53 ***    17.08 ***    19.86 ***
Total panel obs.                2,806        2,753        2,690

                                 (4)          (5)          (6)

Lagged RM 1 (-1)              0.057 ***    0.062 ***     0.050***
                               (0.008)      (0.008)      (0.008)
Lagged inflation (-1)         0.025 ***    0.022 ***     0.024***
                               (0.003)      (0.003)      (0.003)
Lagged GDP growth (-1)          -0.036       -0.016       -0.015
                               (0.058)      (0.059)      (0.059)
Exchange rate regime          0.019 ***    0.019 ***     0.020***
                               (0.003)      (0.003)      (0.003)
Government debt                0.045 *     0.061 ***      0.019
                               (0.023)      (0.022)      (0.024)
Natural resources             0.332 ***    0.295 ***    0.323 ***
                               (0.048)      (0.049)      (0.048)
Shadow economy 1              0.356 ***     0.208 **    0.320 ***
                               (0.085)      (0.087)      (0.089)
Lagged govt turnover (-1)

Military

Lagged constl change (-1)      -0.0004       -0.002       0.000
                               (0.010)      (0.010)      (0.010)
Lagged cabinet change (-1)      -0.003       -0.006       -0.006
                               (0.005)      (0.005)      (0.005)
Lagged exec change (-1)

Financial openness            -0.009 ***     -0.004      -0.006 *
                               (0.003)      (0.003)      (0.003)
Credit to private                            -0.030     -0.198 ***
                                            (0.025)      (0.033)
M2                            0.183 ***                 0.335 ***
                               (0.029)                   (0.040)
1990-1991 dummy                 0.016       0.023 **      0.018
                               (0.011)      (0.012)      (0.012)
2001 dummy                      0.006        0.008        0.002
                               (0.014)      (0.014)      (0.014)
2008-2009 dummy                 -0.022       -0.021       -0.020
                               (0.014)      (0.015)      (0.015)
Quadratic time trend          -0.203 **      -0.056       0.261
                               (0.081)      (0.084)      (0.086)
Constant                      -0.188 ***    -0.066 *    -0.162 ***
                               (0.038)      (0.039)      (0.040)
[R.sup.2] (within)              0.100        0.085        0.109
F-test                        21.77 ***    17.88 ***    21.70 ***
Total panel obs.                3,101        3,048        2,987

                                 (7)          (8)          (9)

Lagged RM 1 (-1)              0.056 ***    0.062 ***    0.050 ***
                               (0.008)      (0.008)      (0.008)
Lagged inflation (-1)         0.025 ***    0.022 ***    0.024 ***
                               (0.003)      (0.003)      (0.003)
Lagged GDP growth (-1)          -0.030       -0.008       -0.008
                               (0.058)      (0.059)      (0.059)
Exchange rate regime          0.019 ***    0.019 ***    0.020 ***
                               (0.003)      (0.003)      (0.003)
Government debt                0.044 *      0.060***      0.018
                               (0.023)      (0.022)      (0.024)
Natural resources             0.332 ***     0.296***    0.324 ***
                               (0.048)      (0.049)      (0.048)
Shadow economy 1              0.357 ***     0.209 **    0.321 ***
                               (0.085)      (0.087)      (0.089)
Lagged govt turnover (-1)

Military

Lagged constl change (-1)       -0.003       -0.005       -0.002
                               (0.010)      (0.010)      (0.010)
Lagged cabinet change (-1)

Lagged exec change (-1)         0.003        0.003        0.001
                               (0.006)      (0.007)      (0.007)
Financial openness            -0.009 ***     -0.004      -0.006 *
                               (0.003)      (0.003)      (0.003)
Credit to private                            -0.028     -0.196 ***
                                            (0.025)      (0.033)
M2                            0.183 ***                 0.334 ***
                               (0.029)                   (0.040)
1990-1991 dummy                 0.016       0.023 *       0.018
                               (0.011)      (0.012)      (0.012)
2001 dummy                      0.006        0.008        0.002
                               (0.014)      (0.014)      (0.014)
2008-2009 dummy                 -0.021       -0.020       -0.019
                               (0.014)      (0.015)      (0.015)
Quadratic time trend          -0.207 **      -0.064     -0.269 ***
                               (0.081)      (0.083)      (0.086)
Constant                      -0.190 ***    -0.069 *    -0.165 ***
                               (0.038)      (0.039)      (0.040)
[R.sup.2] (within)              0.100        0.084        0.109
F-test                        21.75 ***    17.79 ***    21.59 ***
Total panel obs.                3,101        3,048        2,987

Notes: The dependent variable (Seigniorage 1) is measured as
the change in reserve money, divided by government spending.

RM 1 is reserve money (line 14a of IMF-IFS) as percentage of
government spending. Ethnic polarization is omitted because
of collinearity.

* p <.1; ** p < .05; *** p < .01.

REFERENCES

Aisen, A., and F. J. Veiga. "The Political Economy of Seigniorage." Journal of Development Economics, 87(1), 2008, 29-50.

Arellano, M., and S. Bond. "Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations." Review of Economic Studies, 58(2), 1991, 277-97.

Arellano. M., and O. Bover. "Another Look at the Instrumental Variable Estimation of Error-Components Models." Journal of Econometrics, 68(1), 1995, 29-51.

Arnone, M., B. J. Laurens, J. F. Segalotto, and M. Sommer. "Central Bank Autonomy: Lessons from Global Trends." IMF Staff Papers, 56(2), 2009.

Baltagi, B. H. Econometric Analysis of Panel Data. 5th ed. Chichester, UK: John Wiley & Sons, 2013.

Baltagi, B. H., P. O. Demetriades, and S. H. Law. "Financial Development and Openness: Evidence from Panel Data." Journal of Development Economics, 89(2), 2009, 285-96.

Banks, A. S., and K. A. Wilson. "Cross-National Time-Series Data Archive." Databanks International, 2015. Accessed April 22, 2015. http://www .databanksinternational.com/.

Beck, T., G. Clarke, A. Groff, P. Keefer, and P. Walsh. "New Tools in Comparative Political Economy: The Database of Political Institutions." World Bank Economic Review, 15(1), 2001, 165-76.

Besley, T., and T. Persson. "Why Do Developing Countries Tax So Little?" The Journal of Economic Perspectives, 28(4), 2014, 99-120.

Bittencourt, M., R. Gupta, and L. Stander. "Tax Evasion, Financial Development and Inflation: Theory and Empirical Evidence." Journal of Banking & Finance, 41, 2014, 194-208.

Blundell, R., and S. Bond. "Initial Conditions and Moment Restrictions in Dynamic Panel Data Models." Journal of Econometrics, 87(1), 1998, 115-43.

Chinn, M. D., and H. Ito. "What Matters for Financial Development? Capital Controls, Institutions, and Interactions." Journal of Development Economics, 81(1), 2006, 163-92.

Click, R. W. "Seigniorage in a Cross-Section of Countries." Journal of Money, Credit, and Banking, 30(2), 1998, 154-71.

Cukierman, A., S. Edwards, and G. Tabellini. "Seigniorage and Political Instability." American Economic Review, 82(3), 1992, 537-55.

Desmet, K., I. Ortuno-Ortin, and R. Wacziarg. "The Political Economy of Linguistic Cleavages." Journal of Development Economics, 97(2), 2012, 322-38.

Dreher, A., J. E. Sturm, and J. de Haan. "Does High Inflation Cause Central Bankers to Lose Their Job? Evidence Based on a New Data Set." European Journal of Political Economy, 24(4), 2008, 778-87.

Elgin, C., and O. Oztunali. "Shadow Economies Around the World: Model Based Estimates." Bogazici University Department of Economics Working Papers 5, 2012.

Gordon, R., and W. Li. "Tax Structures in Developing Countries: Many Puzzles and a Possible Explanation." Journal of Public Economics, 93(7), 2009, 855-66.

Horowitz, D. L. Ethnic Groups in Conflict. Berkeley: University of California Press, 1985.

Ilzetzki, E., C. M Reinhart, and K. S. Rogoff. "Exchange Rate Arrangements Entering the 21st Century: Which

Anchor Will Hold?" Unpublished Manuscript and Data, 2010. Accessed April 1, 2014. http://personal.lse.ac.uk/ ilzetzki/index.htm/Data.htm.

International Monetary Fund. "Annual Report on Exchange Arrangements and Exchange Restrictions." 2012. Accessed May 10, 2014. http://www.imf.org/external/ pubs/nft/2012/eaer/ar2012.pdf.

--. "International Financial Statistics." 2014. Accessed March 21, 2014. http://elibrary-data.imf .org/QueryBuilder.aspx?key=19784651&s=322.

Jensen, A. "State-Building in Resource-Rich Economies." Atlantic Journal of Economics, 39(2), 2011, 171-93.

Keefer, P. "Database of Political Institutions: Changes and Variable Definitions. The World Bank, Development Research Group." 2012. Accessed March 30, 2014. http://siteresources.worldbank.org/INTRES/ Resources/469232-1107449512766/DPI2012_ Codebook2.pdf.

Levine, R., N. Loayza, and T. Beck. "Financial Intermediation and Growth: Causality and Causes." Journal of Monetary Economics, 46(1), 2000, 31-77.

Montalvo, J. G., and M. Reynal-Querol. "Ethnic Polarization, Potential Conflict, and Civil Wars." American Economic Review, 95(3), 2005a, 796-816.

--. "Ethnic Diversity and Economic Development." Journal of Development Economics, 76(2), 2005b, 293-323.

Mundlak, Y. "On the Pooling of Time Series and Cross Section Data." Econometrica, 46(1), 1978, 69-85.

National Bureau of Economic Research. "US Business Cycle Expansions and Contractions." 2014. Accessed July 25, 2014. http://www.nber.org/cycles/cyclesmain.html.

Nicolini, J. P. "Tax Evasion and the Inflation Optimal Tax." Journal of Development Economics, 55(1), 1998, 215-32.

Roodman, D. "A Note on the Theme of Too Many Instruments." Oxford Bulletin of Economics and Statistics, 71(1), 2009a, 135-58.

--. "How to Do xtabond2: An Introduction to Difference and System GMM in Stata." Stata Journal, 9(1), 2009b. 86.

Roubini, N., and X. Saia-i-Martin. "A Growth Model of Inflation, Tax Evasion, and Financial Repression." Journal of Monetary Economics, 35(2), 1995, 275-301.

Schneider, F., A. Buehn, and C. E. Montenegro. "New Estimates for the Shadow Economies all over the World." International Economic Journal, 24(4), 2010, 443-61.

Taylor, C. L., and M. C. Hudson World Handbook of Political and Social Indicators. 2nd ed. New Haven, CT: Yale University Press, 1972.

Windmeijer, F. "A Finite Sample Correction for the Variance of Linear Efficient Two-Step GMM Estimators." Journal of Econometrics, 126(1), 2005, 25-51.

Woo, J. "Economic, Political, and Institutional Determinants of Public Deficits." Journal of Public Economics, 87(3), 2003, 387-426.

--. "Social Polarization, Fiscal Instability and Growth." European Economic Review, 49(6), 2005, 1451-77.

Wooldridge, J. M. Econometric Analysis of Cross Section and Panel Data. 2nd ed. Cambridge, MA: MIT Press, 2011.

World Bank. "World Development Indicators." 2014a. Accessed May 1, 2014. http://data.worldbank.org/ indicator/all.

--. "Worldwide Governance Indicators." 2014b. Accessed May 30, 2014. http://info.worldbank.org/ governance/wgi.

NASR G. ELBAHNASAWY and MICHAEL A. ELLIS *

* We would like to thank Eric Johnson, Curtis Reynolds, Shawn Rohlin, James Saunoris, Bruce McGough (co-editor), and an anonymous referee for their valuable comments and suggestions that considerably improved the paper. All remaining errors are ours.

Elbahnasawy: Assistant Professor, Department of Economics, Kent State University, Kent, OH 44242; Department of Economics, Faculty of Law, Mansoura University, Mansoura, Egypt. Phone 3306721210, Fax 3306729808, E-mail nelbahn@kent.edu

Ellis: Associate Professor, Department of Economics, Kent State University, Kent, OH 44242. Phone 3306721091, Fax 3306729808, E-mail mellis@kent.edu

(1.) We follow the definition of the shadow economy from Schneider, Buehn, and Montenegro (2010) as "all market-based legal production of goods and services that are deliberately concealed from public authorities."

(2.) For examples, see Cukierman, Edwards, and Tabellini (1992), Click (1998), and Aisen and Veiga (2008).

(3.) See Jensen (2011) and Besley and Persson (2014).

(4.) See Gordon and Li (2009).

(5.) See, for example, Roubini and Sala-i-Martin (1995), Nicolini (1998), and Bittencourt, Gupta, and Stander (2014).

(6.) A list of all countries included in the analysis is available from the authors upon request.

(7.) Government spending is calculated by multiplying nominal GDP from line 99b of IMF Financial Indicators by general government final consumption as a share of GDP, taken from the World Bank World Development Indicators.

(8.) This measure of the shadow economy was recently used by Bittencourt, Gupta, and Stander (2014).

(9.) The MIMIC estimates of the shadow economy cover the period 1999-2007 while the Elgin and Oztunali (2012) measures cover the period 1950-2009. The correlation between these measures is .987, as reported in Table A2.

(10.) Jensen (2011) provides evidence that increased natural resource rents decrease conventional taxation.

(11.) In our sample, the mean of natural resource rents is 1.8% and 9.5% of GDP in OECD (Organization for Economic Co-operation and Development) and non-OECD countries, respectively. See Table A3 for more details.

(12.) The index divides regimes into six categories: fixed, limited flexibility, managed floating, freely floating, freely falling, and dual markets. As a robustness check, we also use the IMF de jure coarse classification of exchange rate regime.

(13.) See International Monetary Fund (2012, 2014).

(14.) The ELF index was originally computed by Taylor and Hudson (1972).

(15.) For detailed discussion, see Horowitz (1985).

(16.) Woo (2003, 2005) found that higher social polarization--proxied by ethnic heterogeneity--results in higher polarization of preferences, which is translated to political parties and interest groups lobbying for various types of government spending and public deficits, leading to greater fiscal and political instability.

(17.) Indirect measures of polarization have been used in previous studies. Aisen and Veiga (2008) use income inequality (proxied by Gini index), ethnic fractionalization, and religious fractionalization as proxies for polarization, while Cukierman, Edwards, and Tabellini (1992) use a dummy variable for democratic regimes as a proxy for political polarization to find that polarization is a major determinant of seigniorage. We use a direct and in our opinion more appropriate measure of polarization in this study, as discussed above.

(18.) The measure of fractionalization is maximized when each individual belongs to a different group, while the measure of polarization is maximized with two groups of equal size. For more details, see Desmet, Ortuno-Ortin, and Wacziarg (2012). We have also used measures of religious polarization and religious fractionalizations (from Montalvo and Reynal-Querol 2005b) to check the robustness of the results.

(19.) For more details on this measure, see Keefer (2012, 20-21).

(20.) See Aisen and Veiga (2008).

(21.) For the methodology used to compute this index, see Amone et al. (2009).

(22.) Dreher, Sturm, and de Haan (2008) argue that the turnover rate of central bank governors is a poor indicator of central bank independence, especially when it comes to evaluating the influence of central bank independence on inflation.

(23.) The inflation rate is measured as the annual growth rate of the GDP deflator.

(24.) Lagged adjustment of the real money stock can result from lagged adjustment of money demand because of adaptive inflation expectations or from lagged adjustment of the real money stock to immediate changes in inflation expectations and money demand.

(25.) See Roodman (2009b).

(26.) Dates of global economic contractions are based on the U.S. business cycle reference dates--from the National Bureau of Economic Research (NBER)--that are determined by the NBER's Business Cycle Dating Committee. A sizeable economic contraction in the United States would likely have global negative economic impact as well. The latest three U.S. economic contractions are dated as December 2007 to June 2009, March 2001 to November 2001, and July 1990 to March 1991. For more details, see NBER (2014).

(27.) It is worth mentioning that Aisen and Veiga (2008) tried to utilize dynamic panel data models to account for endogeneity of explanatory variables, but lagged seigniorage was never statistically significant in their models.

(28.) The existence of the lagged dependent variable among regressors is immediately apparent when we express seigniorage as the difference in reserve money--as in Equation (2)--and that the reserve money term appears on both sides of the equation.

(29.) Additionally, the dynamic panel methodology deals effectively with nonstationarity--the system GMM estimator ensures that all regressors are stationary in any event--and measurement error (Baltagi, Demetriades, and Law 2009).

(30.) The instrument count is now linear in T. For each period, separate instruments are still generated but their number per period is capped at ten.

(31.) Furthermore, the difference-in-Hansen test for the subset based on the dependent variable is statistically insignificant in each specification.

(32.) The coefficient on lagged dependent variable lies within the range for dynamic stability.

(33.) The mean value of the exchange rate regime index in our sample is 2.16 on a six-point scale, where a value of 6 indicates the most flexible exchange rate regime.

(34.) See, for example, Cukierman, Edwards, and Tabellini (1992) and Click (1998).

(35.) It is worth noting that this result does not change when using the turnover rate of central bank governors (de facto central bank independence), contradicting Aisen and Veiga (2008) and Click (1998) findings that high turnover rate of central bank governs significantly increases seigniorage.

(36.) The results are the same when introducing [+ or -] 5% measurement error into estimates of the shadow economy. We have also created a 3-year moving average dataset for the size of the shadow economy--instead of annual values--and the results are still in line with those of the main tables.

(37.) These results can be obtained from the authors on request.

(38.) Including decade dummies in the models of Table 2 also gives similar results and can be obtained from the authors on request.

(39.) Cukierman, Edwards, and Tabellini (1992) cover the period 1971-1982, Click (1998) covers the period 1971-1990, and Aisen and Veiga (2008) cover the period 1970-1999.

(40.) In this table, we are treating Lagged reserve money and Lagged inflation as endogenous variables in the GMM-style instruments (with lag ranges set to seven) to test the robustness of the results to variations in the instrument sets. Treating only Lagged reserve money as the endogenous variable in the GMM-style instruments gives similar results.

(41.) Using 1970-1991 period provides the same conclusion.

(42.) Both measures of inflation are drawn from WB-WD1.

(43.) These results, and several others referred to in this section, are available from the authors upon request to conserve space.

(44.) Countries with poor credit ratings are likely to rely more on seigniorage than countries with good credit ratings (Click 1998). As a proxy for country credit worthiness, we use the average interest rate on new external debt commitments, which represents the interest rate on all new public and publicly guaranteed loans contracted during the year obtained from the WB-WDI (World Bank 2014a). We have also utilized the interest payments on government debt to domestic and foreign residents, from WB-WDI.

(45.) We employed the net barter terms of trade index, drawn from the WB-WDI, constructed as the percentage ratio of the export to the import unit value indexes, calculated relative to the base year 2000.

(46.) The political stability and absence of violence index comes from the World Bank's Worldwide Governance Indicators (World Bank 2014b). It measures perceptions of the probability that a government will be destabilized or overthrown by unconstitutional or violent means.

(47.) Military expenditures as a share of GDP, drawn from the WB-WDI, include all current and capital expenditures on the armed forces.

(48.) With each of these variables, the number of observations drops considerably. We have also explored the possibility of interaction effects between the shadow economy and some of the regressors on seigniorage, but none of them produced useful results.

(49.) See Aisen and Veiga (2008).

(50.) See Cukierman, Edwards, and Tabellini (1992) and Click (1998).

TABLE 1
Estimation Results of the Two-Step System GMM: Seigniorage 1

                         (1)          (2)          (3)          (4)

Lagged RM 1 (-1)      0.104 ***    0.094 ***    0.098 ***    0.098 ***
                       (0.030)      (0.032)      (0.031)      (0.031)
Lagged                  0.008       0.012 *       0.008        0.008
  inflation (-1)       (0.005)      (0.006)      (0.005)      (0.005)
Lagged GDP              -0.057       -0.060       -0.056       -0.051
  growth (-1)          (0.076)      (0.086)      (0.077)      (0.077)
Exchange rate          0.027 **    0.033 ***     0.029 **     0.029 **
  regime               (0.011)      (0.012)      (0.011)      (0.011)
Government debt         0.004        0.005        0.006        0.006
                       (0.030)      (0.030)      (0.032)      (0.033)
Natural resources     0.130 ***    0.123 ***    0.128 ***    0.128 ***
                       (0.028)      (0.029)      (0.028)      (0.029)
Ethnic                  -0.005       0.010        -0.004       -0.004
  polarization         (0.017)      (0.020)      (0.017)      (0.017)
Shadow economy 1      0.113 ***    0.129 ***    0.121 ***    0.120 ***
                       (0.034)      (0.035)      (0.035)      (0.035)
Lagged govt                          -0.006
  turnover (-1)                     (0.008)
Military                             -0.003
                                    (0.010)
Lagged constl                                     -0.011       -0.012
  change (-1)                                    (0.009)      (0.009)
Lagged cabinet                                    -0.005
  change (-1)                                    (0.005)
Lagged exec                                                    -0.001
  change (-1)                                                 (0.007)
CBA

M2

1990-1991 dummy         0.017        0.018        0.020        0.019
                       (0.013)      (0.012)      (0.013)      (0.013)
2001 dummy              0.017        0.017        0.018        0.018
                       (0.015)      (0.014)      (0.015)      (0.015)
2008-2009 dummy       -0.036 **    -0.038 **     -0.034 *     -0.033 *
                       (0.018)      (0.018)      (0.017)      (0.017)
Quadratic time          0.057        0.156        0.054        0.051
  trend                (0.085)      (0.096)      (0.089)      (0.089)
Constant              -0.079 **    -0.104 **    -0.079 **    -0.081 **
                       (0.036)      (0.041)      (0.035)      (0.036)
E-test                   7.64       7.48 ***     6.65 ***     6.72 ***
Arellano-Bond test      0.000        0.000        0.000        0.000
  AR(1), p value
Arellano-Bond test      0.508        0.338        0.542        0.546
  AR(2), p value
Hansen test of          0.224        0.441        0.261        0.273
  overid., p value
Difference-in-           0.26         0.37         0.20         0.27
  Hansen tests,
  p value
Number of                 31           33           33           33
  instruments
Number of                144          144          141          141
  countries
Average obs. per        23.89        21.20        23.75        23.75
  country
Total panel obs.        3,440        3,053        3,349        3,349

                         (5)          (6)          (7)

Lagged RM 1 (-1)       0.094 **    0.098 ***    0.094 ***
                       (0.038)      (0.032)      (0.029)
Lagged                 0.013 **     0.014 **     0.010 *
  inflation (-1)       (0.007)      (0.007)      (0.005)
Lagged GDP              -0.047       -0.084       -0.098
  growth (-1)          (0.095)      (0.083)      (0.073)
Exchange rate          0.030 **    0.032 ***     0.022 *
  regime               (0.013)      (0.012)      (0.011)
Government debt         0.004        -0.005       0.009
                       (0.029)      (0.032)      (0.034)
Natural resources     0.152 ***    0.125 ***    0.131 ***
                       (0.035)      (0.032)      (0.031)
Ethnic                  0.004        -0.002       -0.015
  polarization         (0.025)      (0.019)      (0.017)
Shadow economy 1      0.156 ***    0.157 ***    0.146 ***
                       (0.034)      (0.043)      (0.044)
Lagged govt             -0.004       -0.002
  turnover (-1)        (0.008)      (0.009)
Military                -0.001       -0.001
                       (0.014)      (0.011)
Lagged constl                                     -0.009
  change (-1)                                    (0.009)
Lagged cabinet                                    -0.002
  change (-1)                                    (0.005)
Lagged exec
  change (-1)
CBA                     0.018
                       (0.027)
M2                                  0.044 *      0.047 *
                                    (0.027)      (0.026)
1990-1991 dummy        0.025 *       0.015        0.016
                       (0.014)      (0.012)      (0.013)
2001 dummy              0.023        0.013        0.014
                       (0.015)      (0.015)      (0.015)
2008-2009 dummy       -0.050 ***   -0.046 **    -0.047 **
                       (0.019)      (0.020)      (0.018)
Quadratic time          0.172        0.120        0.038
  trend                (0.107)      (0.100)      (0.089)
Constant              -0.115 **    -0.124 ***   -0.082 **
                       (0.049)      (0.040)      (0.037)
E-test                 9.52 ***     7.04 ***     6.42 ***
Arellano-Bond test      0.000        0.000        0.000
  AR(1), p value
Arellano-Bond test      0.238        0.781        0.977
  AR(2), p value
Hansen test of          0.387        0.418        21.16
  overid., p value
Difference-in-          0.418        0.365         0.22
  Hansen tests,
  p value
Number of                 34           34           34
  instruments
Number of                125          140          137
  countries
Average obs. per        20.58        20.40        23.00
  country
Total panel obs.        2,572        2,856        3,151

Notes: Windmeijer-corrected cluster-robust errors are in
parentheses. The null hypothesis in the Arellano-Bond test
for second-order serial correlation is that there is no
second-order serial correlation in the disturbances, in
differences. The null hypothesis in the Hansen test of over-
identified restrictions is that all instruments are valid.
The null hypothesis in the difference-in-Hansen tests of
exogeneity of the full set of GMM instruments for the levels
equation is that instruments are exogenous. The dependent
variable (Seigniorage 1) is measured as the change in
reserve money, divided by government spending. RM1 is
reserve money (line 14a of IMF-IFS) as percentage of
government spending.

* p <.1; ** p <.05; *** p <.01.

TABLE 2
Estimation Results of the Two-Step System GMM:
Financial Development: Seigniorage 1

                                  (1)          (2)          (3)

Lagged RM 1 (-1)               0.089 ***    0.086 ***     0.074 **
                                 (0.29)      (0.029)      (0.033)
Lagged inflation (-1)           0.016 *       0.012        0.014
                                (0.009)      (0.008)      (0.009)
Lagged GDP growth (-1)           -0.097       -0.039       -0.081
                                (0.086)      (0.090)      (0.087)
Exchange rate regime            0.030 **     0.026 *      0.029 **
                                (0.012)      (0.013)      (0.013)
Government debt                 -0.0004       0.031        -0.028
                                (0.032)      (0.027)      (0.039)
Natural resources              0.140 ***    0.102 ***    0.108 ***
                                (0.035)      (0.034)      (0.037)
Ethnic polarization              -0.002       0.009        -0.019
                                (0.020)      (0.019)      (0.022)
Shadow economy 1               0.157 ***     0.059 *     0.136 ***
                                (0.040)      (0.034)      (0.036)
Lagged govt turnover (-1)        -0.003       -0.006       -0.004
                                (0.009)      (0.008)      (0.009)
Military                         -0.004       -0.011       -0.005
                                (0.012)      (0.011)      (0.012)
Lagged constl change (-1)

Lagged cabinet change (-1)

Lagged exec change (-1)

Financial openness               -0.006       -0.003       -0.003
                                (0.004)      (0.004)      (0.003)
Credit to private                            -0.040 *    -0.187 ***
                                             (0.022)      (0.062)
M2                              0.062 **                 0.232 ***
                                (0.030)                   (0.080)
1990-1991 dummy                  0.014        0.020        0.015
                                (0.012)      (0.013)      (0.013)
2001 dummy                       0.014        0.014        0.011
                                (0.015)      (0.016)      (0.015)
2008-2009 dummy                -0.048 **    -0.056 ***   -0.051 **
                                (0.020)      (0.020)      (0.020)
Quadratic time trend             0.146        0.186        0.082
                                (0.105)      (0.116)      (0.108)
Constant                       -0.123 ***     -0.044     -0.088 **
                                (0.039)      (0.045)      (0.038)
F-test                         11.99 ***    11.78 ***    14.55 ***
Arellano-Bond test AR{1),        0.000        0.000        0.000
  p value
Arellano-Bond test AR(2),        0.619        0.437        0.699
  p value
Hansen test of overid.,          0.354        0.551        0.546
  p value
Difference-in-Hansen tests,      0.331        0.419        0.458
  p value
Number of instruments              35           35           36
Number of countries               139          141          139
Average obs. per country         20.19        19.52        19.35
Total panel obs.                 2,806        2,753        2,690

                                  (4)          (5)          (6)

Lagged RM 1 (-1)               0.093 ***    0.098 ***    0.082 ***
                                (0.027)      (0.030)      (0.030)
Lagged inflation (-1)            0.011        0.009        0.010
                                (0.007)      (0.007)      (0.007)
Lagged GDP growth (-1)           -0.096       -0.041       -0.092
                                (0.077)      (0.080)      (0.076)
Exchange rate regime            0.021 *      0.024 *      0.021 *
                                (0.012)      (0.012)      (0.012)
Government debt                  0.005        0.024        -0.023
                                (0.034)      (0.029)      (0.040)
Natural resources              0.141 ***    0.108 ***    0.114 ***
                                (0.032)      (0.031)      (0.033)
Ethnic polarization              -0.016       -0.003       -0.035
                                (0.019)      (0.017)      (0.021)
Shadow economy 1               0.147 ***     0.061 **    0.131 ***
                                (0.043)      (0.030)      (0.039)
Lagged govt turnover (-1)

Military

Lagged constl change (-1)        -0.012      -0.017 *      -0.014
                                (0.009)      (0.009)      (0.009)
Lagged cabinet change (-1)       -0.002       -0.003       -0.003
                                (0.005)      (0.005)      (0.005)
Lagged exec change (-1)

Financial openness               -0.005       -0.001       -0.001
                                (0.004)      (0.004)      (0.004)
Credit to private                            -0.042 *    -0.198 ***
                                             (0.023)      (0.058)
M2                              0.061 **                 0.239 ***
                                (0.029)                   (0.073)
1990-1991 dummy                  0.015        0.023        0.018
                                (0.014)      (0.014)      (0.014)
2001 dummy                       0.014        0.015        0.012
                                (0.016)      (0.016)      (0.016)
2008-2009 dummy                -0.051 ***   -0.052 ***   -0.052 ***
                                (0.019)      (0.019)      (0.020)
Quadratic time trend             0.067        0.081        -0.003
                                (0.101)      (0.112)      (0.105)
Constant                       -0.088 **      -0.034       -0.059
                                (0.037)      (0.039)      (0.036)
F-test                          9.10 **      8.61 ***    12.90 ***
Arellano-Bond test AR{1),        0.000        0.000        0.000
  p value
Arellano-Bond test AR(2),        0.774        0.555        0.864
  p value
Hansen test of overid.,          0.183        0.290        0.270
  p value
Difference-in-Hansen tests,      0.168        0.293        0.276
  p value
Number of instruments              35           35           36
Number of countries               136          138          136
Average obs. per country         22.80        22.09        21.96
Total panel obs.                 3,101        3,048        2,987

                                  (7)          (8)          (9)

Lagged RM 1 (-1)               0.094 ***    0.099 ***    0.082 ***
                                (0.028)      (0.030)      (0.030)
Lagged inflation (-1)            0.011        0.009        0.010
                                (0.007)      (0.007)      (0.007)
Lagged GDP growth (-1)           -0.095       -0.037       -0.090
                                (0.077)      (0.080)      (0.076)
Exchange rate regime            0.022 *      0.024 *      0.021 *
                                (0.012)      (0.012)      (0.012)
Government debt                  0.005        0.024        -0.023
                                (0.034)      (0.029)      (0.040)
Natural resources              0.141 ***    0.109 ***    0.114 ***
                                (0.032)      (0.031)      (0.033)
Ethnic polarization              -0.016       -0.003       -0.035
                                (0.019)      (0.017)      (0.021)
Shadow economy 1               0.146 ***     0.060 **    0.130 ***
                                (0.043)      (0.030)      (0.038)
Lagged govt turnover (-1)

Military

Lagged constl change (-1)        -0.012      -0.018 *      -0.014
                                (0.009)      (0.009)      (0.009)
Lagged cabinet change (-1)

Lagged exec change (-1)          -0.003       0.0002       -0.003
                                (0.007)      (0.007)      (0.007)
Financial openness               -0.005       -0.001       -0.001
                                (0.004)      (0.004)      (0.004)
Credit to private                            -0.042 *    -0.196 ***
                                             (0.023)      (0.058)
M2                              0.061 **                 0.237 ***
                                (0.029)                   (0.073)
1990-1991 dummy                  0.016        0.023        0.018
                                (0.014)      (0.014)      (0.014)
2001 dummy                       0.014        0.015        0.012
                                (0.016)      (0.016)      (0.016)
2008-2009 dummy                -0.051 ***   -0.052 ***   -0.052 ***
                                (0.019)      (0.019)      (0.020)
Quadratic time trend             0.065        0.077        -0.007
                                (0.101)      (0.113)      (0.105)
Constant                       -0.088 **      -0.036      -0.060 *
                                (0.037)      (0.039)      (0.036)
F-test                          9.01 ***     8.46 ***    12.68 ***
Arellano-Bond test AR{1),        0.000        0.000        0.000
  p value
Arellano-Bond test AR(2),        0.777        0.560        0.866
  p value
Hansen test of overid.,          0.186        0.298        0.275
  p value
Difference-in-Hansen tests,      0.172        0.301        0.283
  p value
Number of instruments              35           35           36
Number of countries               136          138          136
Average obs. per country         22.80        22.09        21.96
Total panel obs.                 3,101        3,048        2,987

Notes: Windmeijer-corrected cluster-robust errors are in
parentheses. The null hypothesis in the Arellano-Bond test
for second-order serial correlation is that there is no
second-order serial correlation in the disturbances, in
differences. The null hypothesis in the Hansen test of over-
identified restrictions is that all instruments are valid.
The null hypothesis in the difference-in-Hansen tests of
exogeneity of the full set of GMM instruments for the levels
equation is that instruments are exogenous. The dependent
variable (Seigniorage 1) is measured as the change in
reserve money, divided by government spending. RM 1 is
reserve money (line 14a of IMF-IFS) as percentage of
government spending.

* p <.1; ** p < .05; *** p <. 01.

TABLE 3
Estimation Results of the Two-Step System GMM: Additional
Controls for Political Instability and Polarization

                                  (1)          (2)          (3)

Lagged RM 1 (-1)               0.093 ***     0.072 **    0.111 ***
                                (0.035)      (0.035)      (0.034)
Lagged inflation (-1)            0.009        0.008       0.013 *
                                (0.006)      (0.006)      (0.007)
Lagged GDP growth (-1)           -0.051       -0.046       -0.086
                                (0.106)      (0.103)      (0.099)
Exchange rate regime           0.042 ***    0.039 ***    0.036' **
                                (0.014)      (0.014)      (0.012)
Government debt                  0.001        0.005        0.007
                                (0.035)      (0.037)      (0.036)
Natural resources              0.144 ***    0.151 ***    0.157 ***
                                (0.034)      (0.037)      (0.035)
Ethnic polarization              0.020        0.016        0.009
                                (0.022)      (0.021)      (0.022)
Shadow economy 1               0.129 ***    0.130 ***    0.135 ***
                                (0.039)      (0.041)      (0.037)
Lagged govt turnover (-1)        0.002                     -0.004
                                (0.008)                   (0.008)
Military                         0.013                     0.0001
                                (0.013)                   (0.011)
Lagged constl change (-1)                     -0.009
                                             (0.012)
Lagged cabinet change (-1)                    -0.002
                                             (0.007)
Years in office                  0.0003       0.001
                                (0.001)      (0.001)
Lawmaking power                                            -0.013
                                                          (0.011)
Legislators frac

Government frac

1990-1991 dummy                  0.011        0.013        0.015
                                (0.012)      (0.013)      (0.012)
2001 dummy                       0.018        0.019        0.024
                                (0.016)      (0.016)      (0.016)
2008-2009 dummy                 -0.029 *     -0.031 *     -0.031 *
                                (0.017)      (0.017)      (0.017)
Quadratic time trend             0.143        0.125        0.152
                                (0.117)      (0.107)      (0.105)
Constant                       -0.136 ***   -0.113 ***   -0.120 ***
                                (0.048)      (0.042)      (0.042)
F-test                          6.84 ***     6.26 ***     7.62 ***
Arellano-Bond test AR(1),        0.000        0.000        0.000
  p value
Arellano-Bond test AR(2),        0.100        0.100        0.203
  p value
Hansen test of overid.,          0.301        0.245        0.453
  p value
Difference-in-Hansen tests,      0.284        0.194        0.351
  p value
Number of instruments              32           32           34
Number of countries               126          124          132
Average obs. per country         19.48        20.43        20.16
Total panel obs.                 2,454        2,533        2,661

                                  (4)          (5)          (6)

Lagged RM 1 (-1)               0.097 ***     0.081 **     0.069 **
                                (0.034)      (0.033)      (0.032)
Lagged inflation (-1)           0.010 *       0.012       0.011 *
                                (0.006)      (0.008)      (0.007)
Lagged GDP growth (-1)           -0.064       -0.032       -0.053
                                (0.098)      (0.092)      (0.095)
Exchange rate regime            0.030 **    0.046 ***    0.036 ***
                                (0.012)      (0.011)      (0.014)
Government debt                  0.020        0.006        0.010
                                (0.040)      (0.033)      (0.042)
Natural resources              0.164 ***    0.133 ***    0.133 ***
                                (0.038)      (0.033)      (0.035)
Ethnic polarization              0.008        0.017        0.008
                                (0.021)      (0.022)      (0.021)
Shadow economy 1               0.124 ***    0.148 ***    0.143 ***
                                (0.038)      (0.038)      (0.044)
Lagged govt turnover (-1)                     -0.001
                                             (0.008)
Military                                      0.003
                                             (0.013)
Lagged constl change (-1)        -0.007                    -0.003
                                (0.011)                   (0.012)
Lagged cabinet change (-1)       -0.007                    -0.008
                                (0.006)                   (0.006)
Years in office

Lawmaking power                  -0.009
                                (0.010)
Legislators frac                              -0.010       -0.015
                                             (0.021)      (0.021)
Government frac

1990-1991 dummy                 0.022 *       0.017       0.021 *
                                (0.013)      (0.011)      (0.012)
2001 dummy                       0.017       0.031 **      0.020
                                (0.016)      (0.015)      (0.015)
2008-2009 dummy                  -0.027       -0.028       -0.030
                                (0.017)      (0.017)      (0.018)
Quadratic time trend             0.148       0.228 *      0.241 *
                                (0.108)      (0.127)      (0.130)
Constant                       -0.093 **    -0.139 ***   -0.094 **
                                (0.039)      (0.042)      (0.044)
F-test                          6.71 ***     6.75 ***     5.61 ***
Arellano-Bond test AR(1),        0.000        0.000        0.000
  p value
Arellano-Bond test AR(2),        0.171        0.119        0.104
  p value
Hansen test of overid.,          0.330        0.516        0.355
  p value
Difference-in-Hansen tests,      0.250        0.462        0.301
  p value
Number of instruments              34           34           34
Number of countries               130          141          138
Average obs. per country         21.02        18.95        19.70
Total panel obs.                 2,733        2,672        2,719

                                  (7)          (8)

Lagged RM 1 (-1)                0.079 **     0.070 *
                                (0.037)      (0.036)
Lagged inflation (-1)           0.012 *      0.011 **
                                (0.007)      (0.005)
Lagged GDP growth (-1)           0.012        0.015
                                (0.098)      (0.099)
Exchange rate regime           0.044 ***     0.034 **
                                (0.013)      (0.014)
Government debt                  0.011        0.015
                                (0.035)      (0.038)
Natural resources              0.134 ***    0.133 ***
                                (0.032)      (0.033)
Ethnic polarization              0.020        0.013
                                (0.024)      (0.022)
Shadow economy 1               0.142 ***    0.128 ***
                                (0.039)      (0.043)
Lagged govt turnover (-1)       -0.0003
                                (0.008)
Military                         0.005
                                (0.013)
Lagged constl change (-1)                     -0.007
                                             (0.012)
Lagged cabinet change (-1)                    -0.007
                                             (0.006)
Years in office

Lawmaking power

Legislators frac

Government frac                  -0.003       -0.008
                                (0.016)      (0.015)
1990-1991 dummy                  0.014        0.014
                                (0.012)      (0.013)
2001 dummy                      0.032 **      0.023
                                (0.015)      (0.014)
2008-2009 dummy                  -0.028      -0.032 *
                                (0.018)      (0.018)
Quadratic time trend             0.183        0.189
                                (0.114)      (0.118)
Constant                       -0.136 ***   -0.094 **
                                (0.044)      (0.043)
F-test                          6.54 ***     5.87 ***
Arellano-Bond test AR(1),        0.000        0.000
  p value
Arellano-Bond test AR(2),        0.083        0.082
  p value
Hansen test of overid.,          0.334        0.311
  p value
Difference-in-Hansen tests,      0.274        0.263
  p value
Number of instruments              32           32
Number of countries               141          138
Average obs. per country         19.15        19.95
Total panel obs.                 2,700        2,753

Notes: Windmeijer-corrected cluster-robust errors are in
parentheses. The null hypothesis in the Arellano-Bond test
for second-order serial correlation is that there is no
second-order serial correlation in the disturbances, in
differences. The null hypothesis in the Hansen test of over-
identified restrictions is that all instruments are valid.
The null hypothesis in the difference in-Hansen tests of
exogeneity of the full set of GMM instruments for the levels
equation is that instruments are exogenous. The dependent
variable (Seigniorage 1) is measured as the change in
reserve money, divided by government spending. RM1 is
reserve money (line 14a of IMF-1FS) as percentage of
government spending.

* p <.1; ** p <.05; *** p <.01.

TABLE 4
Estimation Results of the Two-Step System GMM: Introducing
Noise into Shadow Economy 1 Estimates by [+ or -] 10 %

                         (1)          (2)          (3)          (4)

Lagged RM 1 (-1)      0.105 ***    0.094 ***    0.098 ***    0.098 ***
                       (0.031)      (0.032)      (0.031)      (0.031)
Lagged inflation        0.008       0.013 **     0.009 *      0.009 *
 (-1)                  (0.005)      (0.006)      (0.005)      (0.005)
Lagged GDP growth       -0.058       -0.061       -0.054      -0.049
 (-1)                  (0.075)      (0.086)      (0.076)      (0.077)
Exchange rate          0.027 **    0.033 ***     0.029 **    0.029 **
  regime               (0.011)      (0.012)      (0.011)      (0.011)
Government debt         0.003        0.003        0.003        0.003
                       (0.030)      (0.031)      (0.032)      (0.033)
Natural resources     0.130 ***    0.121 ***    0.127 ***    0.127 ***
                       (0.028)      (0.029)      (0.028)      (0.029)
Ethnic                  -0.003       0.011        -0.002      -0.002
  polarization         (0.018)      (0.021)      (0.017)      (0.017)
Shadow economy 1      0.073 ***    0.085 ***    0.076 ***    0.076 ***
                       (0.028)      (0.029)      (0.028)      (0.028)
Lagged govt                          -0.006
  turnover (-1)                     (0.008)
Military                            -0.0001
                                    (0.011)
Lagged constl                                     -0.010      -0.011
  change (-1)                                    (0.009)      (0.009)
Lagged cabinet                                    -0.004
  change (-1)                                    (0.005)
Lagged exec                                                   -0.001
  change (-1)                                                 (0.007)
CBA

M2

1990-1991 dummy         0.017        0.018        0.020        0.019
                       (0.013)      (0.012)      (0.013)      (0.013)
2001 dummy              0.017        0.016        0.017        0.017
                       (0.014)      (0.014)      (0.015)      (0.014)
2008-2009 dummy       -0.037 **    -0.040 **    -0.035 **    -0.034 **
                       (0.018)      (0.018)      (0.017)      (0.017)
Quadratic time          0.046        0.153        0.042        0.039
  trend                (0.084)      (0.096)      (0.087)      (0.088)
Constant               -0.065 *    -0.089 **     -0.063 *    -0.065 *
                       (0.035)      (0.040)      (0.035)      (0.035)
F-test                 6.69 ***     6.83 ***     5.75 ***    5.78 ***
Arellano-Bond           0.000        0.000        0.000        0.000
  test AR(1),
  p value
Arellano-Bond test      0.513        0.340        0.549        0.553
  AR(2), p value
Hansen test of          0.240        0.437        0.279        0.290
  overid., p value
Difference-in-          0.217        0.355        0.240        0.248
  Hansen tests,
  p value
Number of                 31           33           33          33
  instruments
Number of                144          144          141          141
  countries
Average obs.            23.89        21.20        23.75        23.75
  per country
Total panel obs.        3,440        3,053        3,349        3,349

                         (5)          (6)          (7)

Lagged RM 1 (-1)       0.094 **    0.097 ***    0.094 ***
                       (0.039)      (0.032)      (0.029)
Lagged inflation       0.014 **     0.014 **     0.010 *
 (-1)                  (0.007)      (0.007)      (0.005)
Lagged GDP growth       -0.046       -0.081       -0.093
 (-1)                  (0.094)      (0.083)      (0.072)
Exchange rate          0.030 **    0.034 ***    0.023 ***
  regime               (0.013)      (0.012)      (0.011)
Government debt         0.004        -0.005       0.009
                       (0.029)      (0.031)      (0.033)
Natural resources     0.146 ***    0.119 ***    0.125 ***
                       (0.036)      (0.031)      (0.030)
Ethnic                  0.005        0.003        -0.010
  polarization         (0.026)      (0.019)      (0.017)
Shadow economy 1      0.109 ***    0.098 ***    0.082 ***
                       (0.028)      (0.029)      (0.030)
Lagged govt             -0.004       -0.002
  turnover (-1)        (0.008)      (0.008)
Military                0.002        0.001
                       (0.014)      (0.011)
Lagged constl                                     -0.008
  change (-1)                                    (0.009)
Lagged cabinet                                    -0.002
  change (-1)                                    (0.005)
Lagged exec
  change (-1)
CBA                     0.012
                       (0.028)
M2                                   0.026        0.026
                                    (0.023)      (0.023)
1990-1991 dummy        0.026 *       0.015        0.017
                       (0.014)      (0.012)      (0.013)
2001 dummy              0.021        0.013        0.014
                       (0.015)      (0.015)      (0.015)
2008-2009 dummy       -0.052 ***   -0.046 **    -0.047 **
                       (0.019)      (0.020)      (0.018)
Quadratic time          0.176        0.130        0.039
  trend                (0.107)      (0.100)      (0.087)
Constant               -0.095 *    -0.101 ***    -0.056 *
                       (0.048)      (0.036)      (0.034)
F-test                 8.46 ***     7.23 ***     6.00 ***
Arellano-Bond           0.000        0.000        0.000
  test AR(1),
  p value
Arellano-Bond test      0.227        0.799        0.994
  AR(2), p value
Hansen test of          0.375        0.466        0.318
  overid., p value
Difference-in-          0.431        0.397        0.242
  Hansen tests,
  p value
Number of                 34           34           34
  instruments
Number of                125          140          137
  countries
Average obs.            20.58        20.40        23.00
  per country
Total panel obs.        2,572        2,856        3,151

Notes: Windmeijer-corrected cluster-robust errors are in
parentheses. The null hypothesis in the Arellano-Bond test
for second-order serial correlation is that there is no
second-order serial correlation in the disturbances, in
differences. The null hypothesis in the Hansen test of over-
identified restrictions is that all instruments are valid.
The null hypothesis in the difference in-Hansen tests of
exogeneity of the full set of GMM instruments for the levels
equation is that instruments are exogenous. The dependent
variable (Seigniorage 1) is measured as the change in
reserve money, divided by government spending. KM 1 is
reserve money (line 14a of IMF-IFS) as percentage of
government spending.

* p<.1; ** p < .05; *** p < .01.

TABLE 5
Estimation Results of the Two-Step System GMM: Financial
Development; Introducing Noise into Shadow Economy 1
Estimates by [+ or -] 10%

                                    (1)          (2)          (3)

Lagged RM 1 (-1)                 0.089 ***    0.086 ***     0.073 **
                                  (0.029)      (0.029)      (0.032)
Lagged inflation (-1)             0.016 *       0.012        0.013
                                  (0.009)      (0.008)      (0.008)
Lagged GDP growth (-1)             -0.093       -0.038       -0.078
                                  (0.086)      (0.090)      (0.086)
Exchange rate regime              0.031 **     0.027 **     0.030 **
                                  (0.012)      (0.013)      (0.012)
Government debt                    -0.001       0.030        -0.028
                                  (0.031)      (0.027)      (0.039)
Natural resources                0.132 ***    0.098 ***    0.100 ***
                                  (0.034)      (0.033)      (0.036)
Ethnic polarization                0.003        0.010        -0.016
                                  (0.020)      (0.019)      (0.022)
Shadow economy 1                 0.090 ***      0.027      0.079 ***
                                  (0.029)      (0.028)      (0.029)
Lagged govt turnover (-1)          -0.002       -0.006       -0.004
                                  (0.008)      (0.008)      (0.009)
Military                           -0.003       -0.010       -0.004
                                  (0.012)      (0.011)      (0.012)
Lagged constl change (-1)

Lagged cabinet change (-1)

Lagged exec change (-1)

Financial openness                 -0.006       -0.003       -0.003
                                  (0.004)      (0.004)      (0.003)
Credit to private                             -0.047 **    -0.194 ***
                                               (0.022)      (0.062)
M2                                 0.043                   0.222 ***
                                  (0.027)                   (0.079)
1990-1991 dummy                    0.015        0.020        0.016
                                  (0.012)      (0.013)      (0.013)
2001 dummy                         0.014        0.013        0.011
                                  (0.015)      (0.016)      (0.015)
2008-2009 dummy                  -0.048 **    -0.055 ***   -0.050 **
                                  (0.020)      (0.020)      (0.020)
Quadratic time trend               0.158        0.184        0.088
                                  (0.105)      (0.116)      (0.108)
Constant                         -0.096 ***     -0.031      -0.065 *
                                  (0.035)      (0.042)      (0.035)
F-test                           10.75 ***    11.92 ***    14.95 ***
Arellano-Bond test AR(1), p        0.000        0.000        0.000
  value
Arellano-Bond test AR(2), p        0.641        0.441        0.715
  value
Hansen test of overid.,            0.396        0.558        0.584
  p value
Difference-in-Hansen tests, p      0.352        0.426        0.484
  value
Number of instruments                35           35           36
Number of countries                 139          141          139
Average obs. per country           20.19        19.52        19.35
Total panel obs.                   2,806        2,753        2,690

                                    (4)          (5)          (6)

Lagged RM 1 (-1)                 0.095 ***    0.099 ***    0.083 ***
                                  (0.027)      (0.029)      (0.030)
Lagged inflation (-1)              0.011        0.009        0.010
                                  (0.007)      (0.007)      (0.007)
Lagged GDP growth (-1)             -0.090       -0.039       -0.088
                                  (0.076)      (0.080)      (0.076)
Exchange rate regime              0.022 *      0.024 *      0.022 *
                                  (0.012)      (0.012)      (0.012)
Government debt                    0.005        0.022        -0.023
                                  (0.033)      (0.029)      (0.039)
Natural resources                0.134 ***    0.104 ***    0.106 ***
                                  (0.031)      (0.031)      (0.032)
Ethnic polarization                -0.010       -0.002       -0.030
                                  (0.018)      (0.017)      (0.021)
Shadow economy 1                  0.077 **      0.023       0.068 **
                                  (0.030)      (0.027)      (0.030)
Lagged govt turnover (-1)

Military

Lagged constl change (-1)          -0.011      -0.016 *      -0.012
                                  (0.009)      (0.009)      (0.009)
Lagged cabinet change (-1)         -0.001       -0.003       -0.003
                                  (0.005)      (0.005)      (0.005)
Lagged exec change (-1)

Financial openness                 -0.005       -0.001       -0.002
                                  (0.004)      (0.004)      (0.004)
Credit to private                             -0.051 **    -0.200 ***
                                               (0.023)      (0.057)
M2                                 0.038                   0.220 ***
                                  (0.025)                   (0.069)
1990-1991 dummy                    0.016        0.023        0.018
                                  (0.013)      (0.014)      (0.014)
2001 dummy                         0.014        0.015        0.012
                                  (0.015)      (0.016)      (0.016)
2008-2009 dummy                  -0.051 ***   -0.052 ***   -0.052 ***
                                  (0.019)      (0.019)      (0.020)
Quadratic time trend               0.069        0.075        -0.002
                                  (0.099)      (0.112)      (0.104)
Constant                          -0.060 *      -0.019       -0.034
                                  (0.033)      (0.037)      (0.033)
F-test                            8.26 ***     9.05 ***    13.52 ***
Arellano-Bond test AR(1), p        0.000        0.000        0.000
  value
Arellano-Bond test AR(2), p        0.797        0.560        0.879
  value
Hansen test of overid.,            0.225        0.304        0.311
  p value
Difference-in-Hansen tests, p      0.192        0.298        0.299
  value
Number of instruments                35           35           36
Number of countries                 136          138          136
Average obs. per country           22.80        22.09        21.96
Total panel obs.                   3,101        3,048        2,987

                                    (7)          (8)          (9)

Lagged RM 1 (-1)                 0.095 ***    0.100 ***    0.084 ***
                                  (0.027)      (0.030)      (0.030)
Lagged inflation (-1)              0.011        0.009        0.010
                                  (0.007)      (0.007)      (0.007)
Lagged GDP growth (-1)             -0.090       -0.035       -0.085
                                  (0.076)      (0.080)      (0.076)
Exchange rate regime              0.022 *      0.024 *      0.022 *
                                  (0.012)      (0.012)      (0.012)
Government debt                    0.005        0.022        -0.023
                                  (0.033)      (0.029)      (0.039)
Natural resources                0.134 ***    0.105 ***    0.106 ***
                                  (0.031)      (0.031)      (0.032)
Ethnic polarization                -0.010       -0.002       -0.030
                                  (0.018)      (0.017)      (0.020)
Shadow economy 1                  0.077 **      0.022        0.067"
                                  (0.030)      (0.027)      (0.029)
Lagged govt turnover (-1)

Military

Lagged constl change (-1)          -0.010      -0.017 *      -0.013
                                  (0.009)      (0.009)      (0.009)
Lagged cabinet change (-1)

Lagged exec change (-1)            -0.003       0.000        -0.003
                                  (0.007)      (0.007)      (0.007)
Financial openness                 -0.005       -0.001       -0.001
                                  (0.004)      (0.004)      (0.004)
Credit to private                             -0.050 **    -0.199 ***
                                               (0.023)      (0.057)
M2                                 0.038                   0.219 ***
                                  (0.026)                   (0.069)
1990-1991 dummy                    0.016        0.023        0.018
                                  (0.013)      (0.014)      (0.014)
2001 dummy                         0.014        0.015        0.011
                                  (0.015)      (0.016)      (0.016)
2008-2009 dummy                  -0.051 ***   -0.052 ***   -0.052 ***
                                  (0.019)      (0.019)      (0.020)
Quadratic time trend               0.067        0.071        -0.005
                                  (0.100)      (0.112)      (0.104)
Constant                          -0.060 *      -0.020       -0.035
                                  (0.033)      (0.037)      (0.033)
F-test                            8.19 ***     8.89 ***    13.35 ***
Arellano-Bond test AR(1), p        0.000        0.000        0.000
  value
Arellano-Bond test AR(2), p        0.799        0.564        0.881
  value
Hansen test of overid.,            0.228        0.312        0.315
  p value
Difference-in-Hansen tests, p      0.196        0.306        0.305
  value
Number of instruments                35           35           36
Number of countries                 136          138          136
Average obs. per country           22.80        22.09        21.96
Total panel obs.                   3,101        3,048        2,987

Notes: Windmeijer-corrected cluster-robust errors are in
parentheses. The null hypothesis in the Arellano-Bond test
for second-order serial correlation is that there is no
second-order serial correlation in the disturbances, in
differences. The null hypothesis in the Hansen test of over-
identified restrictions is that all instruments are valid.
The null hypothesis in the difference in-Hansen tests of
exogeneity of the full set of GMM instruments for the levels
equation is that instruments are exogenous. The dependent
variable (Seigniorage 1) is measured as the change in
reserve money, divided by government spending. RM1 is
reserve money (line 14a of IMF-IFS) as percentage of
government spending.

*p < .01; **p < .05; ***p < .01.

TABLE 6
Estimation Results of the Two-Step System GMM: Seigniorage 2

                        (1)          (2)          (3)          (4)

Lagged RM 2 (-1)     0.238 ***     0.222 **    0.225 ***    0.225 ***
                      (0.064)      (0.110)      (0.068)      (0.068)
Lagged               0.002 ***    0.002 ***    0.002 ***    0.002 ***
  inflation (-1)      (0.001)      (0.001)      (0.001)      (0.001)
Lagged GDP             -0.003       -0.005       -0.003       -0.002
  growth (-1)         (0.011)      (0.012)      (0.011)      (0.011)
Exchange rate        0.003 ***    0.003 ***    0.003 ***    0.003 ***
  regime              (0.001)      (0.001)      (0.001)      (0.001)
Government debt        -0.005       -0.004       -0.004       -0.004
                      (0.007)      (0.010)      (0.008)      (0.008)
Natural resources    0.016 ***    0.015 ***    0.015 ***    0.015 ***
                      (0.004)      (0.004)      (0.004)      (0.004)
Ethnic                -0.006 *      -0.004       -0.005       -0.005
  polarization        (0.004)      (0.004)      (0.003)      (0.003)
Shadow economy 1     0.017 ***    0.018 ***    0.016 ***    0.016 ***
                      (0.005)      (0.007)      (0.005)      (0.005)
Lagged govt                         -0.001
  turnover (-1)                    (0.001)
Military                            -0.002
                                   (0.002)
Lagged constl                                  -0.003 **    -0.004 **
  change (-1)                                   (0.001)      (0.001)
Lagged cabinet                                   -0.001
  change (-1)                                   (0.001)
Lagged exec                                                   0.001
  change (-1)                                                (0.001)
CBA

M2

1990-1991 dummy        0.001        0.001        0.001        0.001
                      (0.002)      (0.002)      (0.002)      (0.002)
2001 dummy            0.006 **     0.005 **     0.006 **     0.006 **
                      (0.003)      (0.002)      (0.003)      (0.003)
2008-2009 dummy      -0.006 **    -0.005 **    -0.005 **    -0.005 **
                      (0.002)      (0.002)      (0.002)      (0.002)
Quadratic time         -0.006       0.007        -0.007       -0.008
  trend               (0.013)      (0.021)      (0.014)      (0.014)
Constant             -0.020 ***    -0.022 *    -0.019 ***   -0.019 ***
                      (0.007)      (0.012)      (0.007)      (0.007)
F-test                8.39 ***     7.80 ***     7.00 ***     6.58 ***
Arellano-Bond          0.000        0.000        0.000        0.000
  test AR(1),
  p value
Arellano-Bond          0.126        0.106        0.125        0.128
  test AR(2),
  p value
Hansen test of         0.391        0.194        0.431        0.430
  overid.,
  p value
Difference-in-         0.269        0.287        0.301        0.300
  Hansen tests,
  p value
Number of                18           20           20           20
  instruments
Number of               144          144          141          141
  countries
Average obs.           24.88        21.88        24.69        24.70
  per country
Total panel obs.       3,582        3,151        3,481        3,482

                        (5)          (6)          (7)

Lagged RM 2 (-1)       0.242      0.224 ***    0.233 ***
                      (0.148)      (0.079)      (0.063)
Lagged                0.001 **    0.002 ***    0.002 ***
  inflation (-1)      (0.001)      (0.001)      (0.001)
Lagged GDP             -0.006       -0.009       -0.010
  growth (-1)         (0.016)      (0.011)      (0.010)
Exchange rate         0.003 **    0.004 ***    0.003 ***
  regime              (0.001)      (0.001)      (0.000)
Government debt        -0.007       -0.001       -0.002
                      (0.013)      (0.006)      (0.006)
Natural resources    0.019 ***    0.013 ***    0.014 ***
                      (0.005)      (0.004)      (0.003)
Ethnic                 -0.006       -0.002       -0.004
  polarization        (0.005)      (0.003)      (0.003)
Shadow economy 1      0.023 **     0.012 **     0.011 **
                      (0.009)      (0.005)      (0.005)
Lagged govt            -0.001       -0.001
  turnover (-1)       (0.001)      (0.001)
Military               -0.003       -0.002
                      (0.005)      (0.002)
Lagged constl                                  -0.003 **
  change (-1)                                   (0.001)
Lagged cabinet                                   -0.001
  change (-1)                                   (0.001)
Lagged exec
  change (-1)
CBA                    0.004
                      (0.004)
M2                                  -0.003       -0.002
                                   (0.008)      (0.006)
1990-1991 dummy        0.002        0.001        0.001
                      (0.002)      (0.001)      (0.001)
2001 dummy            0.006 **      0.002       0.005 *
                      (0.002)      (0.002)      (0.003)
2008-2009 dummy      -0.007 **    -0.006 **    -0.007 **
                      (0.003)      (0.002)      (0.003)
Quadratic time         0.005        0.011        -0.002
  trend               (0.026)      (0.013)      (0.011)
Constant               -0.026     -0.019 ***   -0.017 ***
                      (0.017)      (0.004)      (0.004)
F-test                7.59 ***    13.62 ***    12.43 ***
Arellano-Bond          0.000        0.000        0.000
  test AR(1),
  p value
Arellano-Bond          0.082        0.146        0.172
  test AR(2),
  p value
Hansen test of         0.391        0.531        0.554
  overid.,
  p value
Difference-in-         0.521        0.504        0.425
  Hansen tests,
  p value
Number of                21           21           21
  instruments
Number of               125          140          137
  countries
Average obs.           21.25        21.09        23.95
  per country
Total panel obs.       2,656        2,952        3,281

Notes: Windmeijer-corrected cluster-robust errors are in
parentheses. The null hypothesis in the Arellano-Bond test
for second-order serial correlation is that there is no
second-order serial correlation in the disturbances, in
differences. The null hypothesis in the Hansen test of over-
identified restrictions is that all instruments are valid.
The null hypothesis in the difference in-Hansen tests of
exogeneity of the full set of GMM instruments for the levels
equation is that instruments are exogenous. The dependent
variable (Seigniorage 2) is measured as the change in
reserve money, divided by nominal GDR RM 2 is reserve money
(line 14a of IMF-IFS) as percentage of nominal GDR

* p <. 1; p <. 05; p <. 01.

TABLE 7
Estimation Results of the Two-Step System GMM: Financial
Development; Seigniorage 2

                                    (1)          (2)          (3)

Lagged RM 2 (-1)                 0.177 ***     0.179 **     0.176 **
                                  (0.064)      (0.080)      (0.070)
Lagged inflation (-1)             0.002 **     0.001 *      0.002 **
                                  (0.001)      (0.001)      (0.001)
Lagged GDP growth (-1)             -0.008       0.001        -0.006
                                  (0.011)      (0.013)      (0.011)
Exchange rate regime             0.003 ***    0.003 ***    0.004 ***
                                  (0.001)      (0.001)      (0.001)
Government debt                    0.003        0.004        0.003
                                  (0.003)      (0.006)      (0.004)
Natural resources                0.016 ***    0.017 ***    0.016 ***
                                  (0.004)      (0.004)      (0.004)
Ethnic polarization               -0.0004       -0.001      -0.0002
                                  (0.003)      (0.003)      (0.003)
Shadow economy 1                 0.014 ***     0.012 *     0.013 ***
                                  (0.005)      (0.007)      (0.004)
Lagged govt turnover (-1)          -0.001       -0.001       -0.001
                                  (0.001)      (0.001)      (0.001)
Military                           -0.001       -0.002       -0.001
                                  (0.002)      (0.002)      (0.002)
Lagged constl change (-1)

Lagged cabinet change (-1)

Lagged exec change (-1)

Financial openness                 0.0001       0.0001       0.0002
                                  (0.001)      (0.001)      (0.001)
Credit to private                               -0.002       -0.007
                                               (0.003)      (0.013)
M2                                 0.002                     0.006
                                  (0.008)                   (0.021)
1990-1991 dummy                    0.001        0.003        0.002
                                  (0.001)      (0.002)      (0.001)
2001 dummy                         0.001        0.004        0.002
                                  (0.003)      (0.003)      (0.003)
2008-2009 dummy                  -0.006 **    -0.007 ***   -0.007 ***
                                  (0.002)      (0.002)      (0.002)
Quadratic time trend               0.010        0.009        0.004
                                  (0.016)      (0.026)      (0.013)
Constant                         -0.018 ***    -0.016 *    -0.017 ***
                                  (0.003)      (0.009)      (0.005)
F-test                           13.68 ***     7.93 ***    17.52 ***
Arellano-Bond test AR(1), p        0.000        0.000        0.000
  value
Arellano-Bond test AR(2), p        0.117        0.115        0.123
  value
Hansen test of overid.,            0.276        0.147        0.347
  p value
Difference-in-Hansen tests, p      0.336        0.303        0.359
  value
Number of instruments                22           22           23
Number of countries                 139          141          139
Average obs. per country           20.87        20.21        20.04
Total panel obs.                   2.901        2,850        2,786

                                    (4)          (5)          (6)

Lagged RM 2 (-1)                 0.208 ***    0.202 ***    0.205 ***
                                  (0.049)      (0.051)      (0.051)
Lagged inflation (-1)             0.002 **     0.001 **     0.001 **
                                  (0.001)      (0.001)      (0.001)
Lagged GDP growth (-1)             -0.007       0.002        -0.006
                                  (0.010)      (0.011)      (0.010)
Exchange rate regime             0.003 ***    0.003 ***    0.003 ***
                                  (0.001)      (0.001)      (0.001)
Government debt                    0.001        0.002        0.001
                                  (0.004)      (0.005)      (0.003)
Natural resources                0.016 ***    0.016 ***    0.016 ***
                                  (0.003)      (0.004)      (0.003)
Ethnic polarization                -0.003       -0.004       -0.003
                                  (0.003)      (0.003)      (0.003)
Shadow economy 1                 0.013 ***     0.013 **      0.012"
                                  (0.004)      (0.005)      (0.005)
Lagged govt turnover (-1)

Military

Lagged constl change (-1)        -0.003 ***   -0.004 ***   -0.004 ***
                                  (0.001)      (0.001)      (0.001)
Lagged cabinet change (-1)        -0.0002       -0.001       -0.000
                                  (0.001)      (0.001)      (0.001)
Lagged exec change (-1)

Financial openness                 0.001        0.001        0.001
                                  (0.001)      (0.001)      (0.001)
Credit to private                               -0.002       -0.005
                                               (0.003)      (0.009)
M2                                 -0.001                    0.002
                                  (0.006)                   (0.014)
1990-1991 dummy                    0.001        0.002        0.002
                                  (0.001)      (0.002)      (0.001)
2001 dummy                         0.004       0.006 **      0.004
                                  (0.003)      (0.003)      (0.003)
2008-2009 dummy                  -0.006 **    -0.008 ***   -0.008 ***
                                  (0.003)      (0.002)      (0.003)
Quadratic time trend               -0.006       -0.008       -0.009
                                  (0.012)      (0.017)      (0.012)
Constant                         -0.016 ***   -0.015 **    -0.015 ***
                                  (0.003)      (0.006)      (0.004)
F-test                           12.20 ***     6.66 ***    13.97 ***
Arellano-Bond test AR(1), p        0.000        0.000        0.000
  value
Arellano-Bond test AR(2), p        0.142        0.129        0.148
  value
Hansen test of overid.,            0.438        0.392        0.462
  p value
Difference-in-Hansen tests, p      0.313        0.321        0.327
  value
Number of instruments                22           22           23
Number of countries                 136          138          136
Average obs. per country           23.71        23.00        22.88
Total panel obs.                   3,225        3,174        3,112

                                    (7)          (8)          (9)

Lagged RM 2 (-1)                 0.208 ***    0.203 ***    0.205 ***
                                  (0.049)      (0.051)      (0.052)
Lagged inflation (-1)             0.002 **     0.001 **     0.001 **
                                  (0.001)      (0.001)      (0.001)
Lagged GDP growth (-1)             -0.007       0.003        -0.006
                                  (0.010)      (0.011)      (0.010)
Exchange rate regime             0.003 ***    0.003 ***    0.003 ***
                                  (0.001)      (0.001)      (0.001)
Government debt                    0.001        0.002        0.001
                                  (0.004)      (0.005)      (0.003)
Natural resources                0.016 ***    0.017 ***    0.016 ***
                                  (0.003)      (0.004)      (0.003)
Ethnic polarization                -0.003       -0.004       -0.003
                                  (0.003)      (0.003)      (0.003)
Shadow economy 1                 0.013 ***     0.013 **     0.012 **
                                  (0.004)      (0.005)      (0.005)
Lagged govt turnover (-1)

Military

Lagged constl change (-1)        -0.004 ***   -0.004 ***   -0.004 ***
                                  (0.001)      (0.001)      (0.001)
Lagged cabinet change (-1)

Lagged exec change (-1)            0.001        0.001        0.001
                                  (0.001)      (0.001)      (0.001)
Financial openness                 0.001        0.001        0.001
                                  (0.001)      (0.001)      (0.001)
Credit to private                               -0.003       -0.005
                                               (0.003)      (0.009)
M2                                 -0.001                    0.002
                                  (0.006)                   (0.014)
1990-1991 dummy                    0.001        0.002        0.002
                                  (0.001)      (0.002)      (0.001)
2001 dummy                         0.004        0.006"       0.004
                                  (0.003)      (0.003)      (0.003)
2008-2009 dummy                  -0.006 **    -0.007 ***   -0.008 ***
                                  (0.003)      (0.002)      (0.003)
Quadratic time trend               -0.006       -0.009       -0.010
                                  (0.012)      (0.017)      (0.012)
Constant                         -0.016 ***   -0.016 **    -0.015 ***
                                  (0.003)      (0.006)      (0.004)
F-test                           12.05 ***     6.51 ***    13.76 ***
Arellano-Bond test AR(1), p        0.000        0.000        0.000
  value
Arellano-Bond test AR(2), p        0.144        0.132        0.152
  value
Hansen test of overid.,            0.429        0.393        0.453
  p value
Difference-in-Hansen tests, p      0.305        0.319        0.319
  value
Number of instruments                22           22           23
Number of countries                 136          138          136
Average obs. per country           23.72        23.01        22.89
Total panel obs.                   3,226        3,175        3,113

Notes: Windmeijer-corrected cluster-robust errors are in
parentheses. The null hypothesis in the Arellano-Bond test
for second-order serial correlation is that there is no
second-order serial correlation in the disturbances, in
differences. The null hypothesis in the Hansen test of over-
identified restrictions is that all instruments are valid.
The null hypothesis in the difference-in-Hansen tests of
exogeneity of the full set of GMM instruments for the levels
equation is that instruments are exogenous. The dependent
variable (Seigniorage 2) is measured as the change in
reserve money, divided by nominal GDP. RM 2 is reserve money
(line 14a of IMF-IFS) as percentage of nominal GDP.

* p <. 1; ** p < .05; *** p <.01.

TABLE 8
Estimation Results of the Two-Step System GMM: Seigniorage
1, Using Shadow Economy 2

                         (1)          (2)          (3)          (4)

Lagged RM 1 (-1)      0.126 ***    0.127 ***    0.115 ***    0.114 ***
                       (0.027)      (0.027)      (0.028)      (0.028)
Lagged                  -0.010       -0.013       -0.008       -0.008
  inflation (-1)       (0.023)      (0.025)      (0.042)      (0.042)
Lagged GDP              -0.008       0.006        0.003        0.018
  growth (-1)          (0.151)      (0.149)      (0.159)      (0.158)
Exchange rate         0.083 ***    0.086 ***    0.080 ***    0.081 ***
  regime               (0.021)      (0.022)      (0.022)      (0.022)
Government debt        -0.070 *     -0.070 *     -0.069 *     -0.067 *
                       (0.037)      (0.037)      (0.039)      (0.040)
Natural resources     0.116 ***    0.121 ***    0.126 ***    0.128 ***
                       (0.035)      (0.035)      (0.035)      (0.035)
Ethnic                 0.054 *      0.056 *      0.055 *      0.054 *
  polarization         (0.029)      (0.030)      (0.028)      (0.028)
Shadow economy 2      0.195 ***    0.210 ***    0.198 ***    0.199 ***
                       (0.061)      (0.064)      (0.063)      (0.064)
Lagged govt                          -0.001
  turnover (-1)                     (0.015)
Military                             -0.004
                                    (0.017)
Lagged constl                                     0.002        0.001
  change (-1)                                    (0.023)      (0.023)
Lagged cabinet                                   -0.0003
  change (-1)                                    (0.009)
Lagged exec                                                    0.011
  change (-1)                                                 (0.013)
CBA

M2

2001 dummy              0.020        0.024        0.023        0.023
                       (0.016)      (0.016)      (0.016)      (0.016)
Quadratic time        0.893 ***    0.860 ***    0.937 ***    0.952 ***
  trend                (0.285)      (0.288)      (0.288)      (0.289)
Constant              -0.363 ***   -0.372 ***   -0.356 ***   -0.360 ***
                       (0.073)      (0.077)      (0.075)      (0.077)
F-test                10.34 ***     8.92 ***     8.09 ***     7.97 ***
Arellano-Bond test      0.000        0.000        0.000        0.000
  AR(1), p value
Arellano-Bond test      0.158        0.150        0.153        0.149
  AR(2), p value
Hansen test of          0.434        0.410        0.460        0.448
  overid., p value
Difference-in-          0.809        0.790        0.698        0.691
  Hansen tests,
  p value
Number of                 29           31           31           31
  instruments
Number of                143          143          140          140
  countries
Average obs.             7.87         7.78         7.86         7.86
  per country
Total panel obs.        1,126        1,112        1,101        1,101

                         (5)          (6)          (7)

Lagged RM 1 (-1)      0.117 ***    0.128 ***    0.116 ***
                       (0.038)      (0.028)      (0.030)
Lagged                  -0.012       -0.006       0.0002
  inflation (-1)       (0.059)      (0.023)      (0.040)
Lagged GDP              0.021        -0.029       -0.036
  growth (-1)          (0.133)      (0.154)      (0.159)
Exchange rate         0.100 ***    0.069 ***    0.064 ***
  regime               (0.021)      (0.021)      (0.021)
Government debt        -0.064 *     -0.073 *      -0.060
                       (0.037)      (0.039)      (0.039)
Natural resources     0.114 ***    0.119 ***    0.123 ***
                       (0.042)      (0.035)      (0.034)
Ethnic                 0.081 **      0.041        0.040
  polarization         (0.036)      (0.027)      (0.025)
Shadow economy 2      0.196 ***     0.159 **     0.160 **
                       (0.065)      (0.067)      (0.065)
Lagged govt             0.009        0.003
  turnover (-1)        (0.014)      (0.015)
Military                -0.003       -0.010
                       (0.023)      (0.017)
Lagged constl                                     0.002
  change (-1)                                    (0.023)
Lagged cabinet                                    0.002
  change (-1)                                    (0.009)
Lagged exec
  change (-1)
CBA                     0.023
                       (0.044)
M2                                   -0.022       -0.011
                                    (0.040)      (0.037)
2001 dummy             0.037 **      0.016        0.015
                       (0.018)      (0.016)      (0.016)
Quadratic time        1.395 ***    0.820 ***    0.913 ***
  trend                (0.276)      (0.309)      (0.318)
Constant              -0.489 ***   -0.294 ***   -0.295 ***
                       (0.070)      (0.073)      (0.072)
F-test                 9.73 ***     8.98 ***     8.26 ***
Arellano-Bond test      0.000        0.000        0.000
  AR(1), p value
Arellano-Bond test      0.330        0.169        0.159
  AR(2), p value
Hansen test of          0.231        0.485        0.497
  overid., p value
Difference-in-          0.628        0.777        0.705
  Hansen tests,
  p value
Number of                 32           32           32
  instruments
Number of                124          132          130
  countries
Average obs.             7.85         7.64         7.74
  per country
Total panel obs.         973         1,008        1,006

Notes: Windmeijer-corrected cluster-robust errors are in
parentheses. The null hypothesis in the Arellano-Bond test
for second-order serial correlation is that there is no
second-order serial correlation in the disturbances, in
differences. The null hypothesis in the Hansen test of over-
identified restrictions is that all instruments are valid.
The null hypothesis in the difference-in-Hansen tests of
exogeneity of the full set of GMM instruments for the levels
equation is that instruments are exogenous. The dependent
variable (Seigniorage 1) is measured as the change in
reserve money, divided by government spending. RM 1 is
reserve money (line 14a of IMF-IFS) as percentage of
government spending.

* p <.1; ** p < .05; *** p <.01.

TABLE 9
Estimation Results of the Two-Step System GMM: Introducing
Noise into Shadow Economy 2 Estimates by [+ or -] 10%

                         (1)          (2)          (3)          (4)

Lagged RM 1 (-1)      0.123 ***    0.124 ***    0.111 ***    0.111 ***
                       (0.028)      (0.028)      (0.029)      (0.028)
Lagged                  -0.006       -0.009       -0.004       -0.003
  inflation (-1)       (0.023)      (0.025)      (0.044)      (0.044)
Lagged GDP              -0.007       0.006        0.006        0.017
  growth (-1)          (0.151)      (0.150)      (0.157)      (0.157)
Exchange rate         0.084 ***    0.087 ***    0.081 ***    0.081 ***
  regime               (0.022)      (0.023)      (0.022)      (0.023)
Government debt        -0.069 *     -0.069 *     -0.070 *      -0.067
                       (0.039)      (0.040)      (0.041)      (0.042)
Natural resources     0.119 ***    0.122 ***    0.127 ***    0.128 ***
                       (0.035)      (0.035)      (0.035)      (0.035)
Ethnic                 0.054 *      0.056 **     0.056 *      0.055 *
  polarization         (0.030)      (0.032)      (0.029)      (0.029)
Shadow economy 2      0.128 ***    0.136 ***     0.126 **     0.126 **
                       (0.047)      (0.049)      (0.049)      (0.049)
Lagged govt                          -0.002
  turnover (-1)                     (0.015)
Military                             0.001
                                    (0.017)
Lagged constl                                     0.007        0.005
  change (-1)                                    (0.022)      (0.023)
Lagged cabinet                                    0.001
  change (-1)                                    (0.009)
Lagged exec                                                    0.010
  change (-1)                                                 (0.013)
CBA

M2

2001 dummy              0.019        0.022        0.022        0.022
                       (0.016)      (0.016)      (0.016)      (0.016)
Quadratic time        0.867 ***    0.840 ***    0.917 ***    0.928 ***
  trend                (0.289)      (0.291)      (0.292)      (0.292)
Constant              -0.336 ***   -0.345 ***   -0.330 ***   -0.331 ***
                       (0.072)      (0.075)      (0.072)      (0.073)
F-test                10.33 ***     8.76 ***     7.99 ***     7.92 ***
Arellano-Bond test      0.000        0.000        0.000        0.000
  AR(1), p value
Arellano-Bond test      0.162        0.154        0.156        0.152
  AR(2), p value
Hansen test of          0.366        0.344        0.415        0.405
  overid., p value
Difference-in-          0.838        0.808        0.747        0.738
  Hansen tests,
  p value
Number of                 29           31           31           31
  instruments
Number of                143          143          140          140
  countries
Average obs.             7.87         7.78         7.86         7.86
  per country
Total panel obs.        1,126        1,112        1,101        1,101

                         (5)          (6)          (7)

Lagged RM 1 (-1)      0.111 ***    0.127 ***    0.114 ***
                       (0.038)      (0.029)      (0.030)
Lagged                  -0.003       -0.006       -0.002
  inflation (-1)       (0.063)      (0.024)      (0.042)
Lagged GDP              0.031        -0.032       -0.038
  growth (-1)          (0.136)      (0.154)      (0.158)
Exchange rate         0.102 ***    0.070 ***    0.066 ***
  regime               (0.022)      (0.022)      (0.022)
Government debt         -0.064      -0.071 *      -0.058
                       (0.039)      (0.040)      (0.039)
Natural resources      0.110 **    0.116 ***    0.118 ***
                       (0.043)      (0.034)      (0.034)
Ethnic                 0.084 **      0.044       0.043 *
  polarization         (0.038)      (0.027)      (0.025)
Shadow economy 2       0.120 **     0.098 *      0.095 *
                       (0.051)      (0.052)      (0.052)
Lagged govt             0.009        0.003
  turnover (-1)        (0.014)      (0.015)
Military                0.002        -0.010
                       (0.024)      (0.017)
Lagged constl                                     0.004
  change (-1)                                    (0.023)
Lagged cabinet                                    0.003
  change (-1)                                    (0.009)
Lagged exec
  change (-1)
CBA                     0.013
                       (0.047)
M2                                   -0.040       -0.030
                                    (0.037)      (0.035)
2001 dummy             0.036 **      0.016        0.015
                       (0.018)      (0.016)      (0.016)
Quadratic time        1.373 ***    0.844 ***    0.938 ***
  trend                (0.279)      (0.315)      (0.326)
Constant              -0.456 ***   -0.272 ***   -0.270 ***
                       (0.069)      (0.071)      (0.071)
F-test                 9.12 ***     8.71 ***     7.90 ***
Arellano-Bond test      0.000        0.000        0.000
  AR(1), p value
Arellano-Bond test      0.322        0.179        0.169
  AR(2), p value
Hansen test of          0.193        0.460        0.470
  overid., p value
Difference-in-          0.663        0.820        0.768
  Hansen tests,
  p value
Number of                 32           32           32
  instruments
Number of                124          132          130
  countries
Average obs.             7.85         7.64         7.74
  per country
Total panel obs.         973         1,008        1,006

Notes: Windmeijer-corrected cluster-robust errors are in
parentheses. The null hypothesis in the Arellano-Bond test
for second-order serial correlation is that there is no
second-order serial correlation in the disturbances, in
differences. The null hypothesis in the Hansen test of over-
identified restrictions is that all instruments are valid.
The null hypothesis in the difference-in-Hansen tests of
exogeneity of the full set of GMM instruments for the levels
equation is that instruments are exogenous. The dependent
variable (Seigniorage 1) is measured as the change in
reserve money, divided by government spending. RM1 is
reserve money (line 14a of IMF-IFS) as percentage of
government spending.

* p <.1; ** p <.05; *** p <.01.

TABLE 10
Estimation Results of the Two-Step System GMM:
Seigniorage 2, Using Shadow Economy 2

                         (1)          (2)          (3)          (4)

Lagged RM2(-1)         0.174 **     0.189 **     0.108 **     0.108 **
                       (0.078)      (0.077)      (0.042)      (0.042)
Lagged                  0.004        0.003        0.010        0.010
  inflation (-1)       (0.004)      (0.004)      (0.007)      (0.007)
Lagged GDP              -0.011       -0.012       -0.010       -0.007
  growth (-1)          (0.020)      (0.019)      (0.020)      (0.020)
Exchange rate           0.002        0.002        0.001        0.001
  regime               (0.002)      (0.002)      (0.001)      (0.001)
Government debt         -0.003       -0.005       0.002        0.002
                       (0.010)      (0.010)      (0.006)      (0.006)
Natural resources      0.015 **     0.016 **    0.012 ***    0.012 ***
                       (0.006)      (0.006)      (0.004)      (0.004)
Ethnic                  0.001        0.000        0.003        0.003
  polarization         (0.005)      (0.005)      (0.003)      (0.003)
Shadow economy 2       0.021 **     0.022 **     0.017 **     0.017 **
                       (0.009)      (0.009)      (0.007)      (0.007)
Lagged govt                          -0.001
  turnover (-1)                     (0.002)
Military                             -0.001
                                    (0.003)
Lagged constl                                     0.001        0.0003
  change (-1)                                    (0.003)      (0.003)
Lagged cabinet                                    0.0001
  change (-1)                                    (0.001)
Lagged exec                                                    0.002
  change (-1)                                                 (0.002)
CBA

M2

2001 dummy             0.005 *      0.005 *       0.004        0.004
                       (0.003)      (0.002)      (0.003)      (0.003)
Quadratic time         0.095 **     0.095 **    0.113 ***    0.113 ***
  trend                (0.044)      (0.044)      (0.034)      (0.035)
Constant              -0.028 ***   -0.030 ***   -0.022 ***   -0.022 ***
                       (0.010)      (0.010)      (0.008)      (0.008)
F-test                 4.88 ***     4.24 ***     3.84 ***     4.54 ***
Arellano-Bond test      0.000        0.000        0.000        0.000
  AR(1). p value
Arellano-Bond test      0.135        0.185        0.136        0.137
  AR(2), p value
Hansen test of          0.226        0.291        0.178        0.176
  overid., p value
Difference-in-          0.208        0.243        0.241        0.220
  Hansen tests,
  p value
Number of                 16           18           18           18
  instruments
Number of                144          144          141          141
  countries
Average obs.             8.08         7.94         8.08         8.08
  per country
Total panel obs.        1,164        1,144        1,139        1,139

                         (5)          (6)          (7)

Lagged RM2(-1)        0.125 ***    0.143 ***     0.094 **
                       (0.044)      (0.052)      (0.043)
Lagged                  0.009        0.005        0.009
  inflation (-1)       (0.006)      (0.005)      (0.007)
Lagged GDP              -0.007       -0.003       -0.004
  growth (-1)          (0.018)      (0.018)      (0.018)
Exchange rate           0.002       0.002 *       0.001
  regime               (0.001)      (0.001)      (0.001)
Government debt         -0.002       -0.001       0.001
                       (0.005)      (0.004)      (0.004)
Natural resources     0.016 ***     0.012 **     0.012 **
                       (0.005)      (0.005)      (0.005)
Ethnic                  0.001        0.003        0.004
  polarization         (0.003)      (0.003)      (0.003)
Shadow economy 2      0.021 ***     0.019 *      0.021 **
                       (0.008)      (0.010)      (0.010)
Lagged govt             -0.001       -0.001
  turnover (-1)        (0.002)      (0.002)
Military               -0.0003      -0.0002
                       (0.003)      (0.003)
Lagged constl                                     0.001
  change (-1)                                    (0.003)
Lagged cabinet                                    0.001
  change (-1)                                    (0.001)
Lagged exec
  change (-1)
CBA                    0.008 *
                       (0.004)
M2                                   0.003        0.008
                                    (0.009)      (0.008)
2001 dummy              0.004       0.004 *       0.003
                       (0.002)      (0.002)      (0.002)
Quadratic time        0.149 ***    0.101 ***    0.104 ***
  trend                (0.035)      (0.036)      (0.035)
Constant              -0.033 ***   -0.026 ***   -0.024 ***
                       (0.009)      (0.007)      (0.007)
F-test                 5.43 ***     5.33 ***     4.41 ***
Arellano-Bond test      0.000        0.000        0.000
  AR(1). p value
Arellano-Bond test      0.220        0.177        0.126
  AR(2), p value
Hansen test of          0.246        0.264        0.254
  overid., p value
Difference-in-          0.385        0.187        0.290
  Hansen tests,
  p value
Number of                 19           19           19
  instruments
Number of                125          133          131
  countries
Average obs.             7.98         7.81         7.96
  per country
Total panel obs.         998         1,039        1,043

Notes: Windmeijer-corrected cluster-robust errors are in
parentheses. The null hypothesis in the Arellano-Bond test
for second-order serial correlation is that there is no
second-order serial correlation in the disturbances, in
differences. The null hypothesis in the Hansen test of over-
identified restrictions is that all instruments are valid.
The null hypothesis in the difference-in-Hansen tests of
exogeneity of the full set of GMM instruments for the levels
equation is that instruments are exogenous. The dependent
variable (Seigniorage 2) is measured as the change in
reserve money, divided by nominal GDP. RM 2 is reserve money
(line 14a of IMF-1FS) as percentage of nominal GDP.

* p <.1; ** p <.05; *** p <.01.

TABLE 11
Estimation Results of the Two-Step System GMM for Table 1
Models, Using Decade Dummies

                         (1)          (2)          (3)          (4)

Lagged RM 1 (-1)      0.101 ***    0.091 ***    0.095 ***    0.094 ***
                       (0.030)      (0.031)      (0.031)      (0.031)
Lagged                  0.008       0.013 **     0.009 *       0.009
  inflation (-1)       (0.005)      (0.006)      (0.005)      (0.005)
Lagged GDP              -0.061       -0.060       -0.063       -0.058
  growth (-1)          (0.075)      (0.083)      (0.076)      (0.077)
Exchange rate          0.025 **     0.030 **     0.026 **     0.027 **
  regime               (0.011)      (0.012)      (0.011)      (0.012)
Government debt         0.011        0.012        0.012        0.012
                       (0.030)      (0.030)      (0.032)      (0.033)
Natural resources     0.128 ***    0.122 ***    0.125 ***    0.126 ***
                       (0.029)      (0.031)      (0.029)      (0.029)
Ethnic                  -0.001       0.012        0.001        0.001
  polarization         (0.018)      (0.020)      (0.017)      (0.017)
Shadow economy 1      0.103 ***    0.123 ***    0.114 ***    0.113 ***
                       (0.034)      (0.035)      (0.035)      (0.035)
Lagged govt                          -0.006
  turnover (-1)                     (0.008)
Military                             -0.005
                                    (0.011)
Lagged constl                                     -0.011       -0.012
  change (-1)                                    (0.008)      (0.009)
Lagged cabinet                                    -0.005
  change (-1)                                    (0.005)
Lagged exec                                                   -0.0004
  change (-1)                                                 (0.007)
CBA

M2

1980s dummy           -0.025 **    -0.031 **    -0.027 ***   -0.027 ***
                       (0.010)      (0.013)      (0.010)      (0.010)
1990s dummy            -0.020 *     -0.027 *     -0.021 *    -0.022 **
                       (0.011)      (0.014)      (0.011)      (0.011)
2000-2007 dummy         -0.017       -0.022       -0.017       -0.017
                       (0.011)      (0.015)      (0.011)      (0.011)
2008-2009 dummy       -0.050 ***   -0.054 **    -0.048 ***   -0.048 ***
                       (0.018)      (0.021)      (0.018)      (0.018)
Constant                -0.048       -0.057       -0.047       -0.049
                       (0.034)      (0.039)      (0.034)      (0.034)
f-test                 7.12 ***     7.23 ***     6.29 ***     6.34 ***
Arellano-Bond test      0.000        0.000        0.000        0.000
  AR(1), p value
Arellano-Bond test      0.464        0.313        0.499        0.501
  AR(2), p value
Hansen test of          0.257        0.537        0.285        0.295
  overid., p value
Difference-in-          0.236        0.431        0.252        0.258
  Hansen tests,
  p value
Number of                 31           33           33           33
  instruments
Number of                144          144          141          141
  countries
Average obs.            23.89        21.20        23.75        23.75
  per country
Total panel obs.        3,440        3,053        3,349        3,349

                         (5)          (6)          (7)

Lagged RM 1 (-1)       0.095 **    0.091 ***    0.089 ***
                       (0.037)      (0.032)      (0.030)
Lagged                 0.013 **     0.015 **     0.011 *
  inflation (-1)       (0.006)      (0.007)      (0.006)
Lagged GDP              -0.050       -0.094       -0.107
  growth (-1)          (0.094)      (0.082)      (0.073)
Exchange rate          0.028 **     0.029 **     0.019 *
  regime               (0.013)      (0.012)      (0.012)
Government debt         0.010        0.003        0.015
                       (0.029)      (0.032)      (0.035)
Natural resources     0.153 ***    0.121 ***    0.128 ***
                       (0.037)      (0.033)      (0.031)
Ethnic                  0.005        0.000        -0.011
  polarization         (0.025)      (0.020)      (0.018)
Shadow economy 1      0.154 ***    0.154 ***    0.141 ***
                       (0.034)      (0.044)      (0.045)
Lagged govt             -0.005       -0.003
  turnover (-1)        (0.008)      (0.008)
Military                -0.002       -0.002
                       (0.014)      (0.012)
Lagged constl                                     -0.008
  change (-1)                                    (0.009)
Lagged cabinet                                    -0.002
  change (-1)                                    (0.005)
Lagged exec
  change (-1)
CBA                     0.013
                       (0.027)
M2                                  0.047 *      0.051 *
                                    (0.027)      (0.027)
1980s dummy            -0.032 *    -0.033 ***   -0.026 **
                       (0.016)      (0.013)      (0.010)
1990s dummy             -0.021     -0.037 ***   -0.025 **
                       (0.018)      (0.014)      (0.011)
2000-2007 dummy         -0.027       -0.023       -0.015
                       (0.018)      (0.015)      (0.011)
2008-2009 dummy       -0.066 ***   -0.067 ***   -0.062 ***
                       (0.023)      (0.023)      (0.019)
Constant                -0.067      -0.072 *      -0.052
                       (0.048)      (0.040)      (0.036)
f-test                10.27 ***     6.46 ***     5.98 ***
Arellano-Bond test      0.000        0.000        0.000
  AR(1), p value
Arellano-Bond test      0.210        0.759        0.950
  AR(2), p value
Hansen test of          0.481        0.482        0.286
  overid., p value
Difference-in-          0.479        0.398        0.220
  Hansen tests,
  p value
Number of                 34           34           34
  instruments
Number of                125          140          137
  countries
Average obs.            20.58        20.40        23.00
  per country
Total panel obs.        2,572        2.856        3,151

Notes: Windmeijer-corrected cluster-robust errors are in
parentheses. The null hypothesis in the Arellano-Bond test
for second-order serial correlation is that there is no
second-order serial correlation in the disturbances, in
differences. The null hypothesis in the Hansen test of over-
identified restrictions is that all instruments are valid.
The null hypothesis in the difference-in-Hansen tests of
exogeneity of the full set of GMM instruments for the levels
equation is that instruments are exogenous. The dependent
variable (Seigniorage 1) is measured as the change in
reserve money, divided by government spending. RM1 is
reserve money (line 14a of IMF-IFS) as percentage of
government spending.

* p <.1; ** p <.05; *** p <.0l.

TABLE 12
Estimation Results of the Two-Step System GMM, Using Data
over the Period 1970-1999

                         (1)          (2)          (3)          (4)

Lagged RM 1 (-1)      0.159 ***    0.179 ***    0.164 ***    0.164 ***
                       (0.047)      (0.054)      (0.048)      (0.048)
Lagged                  0.008      0.017 ***      0.009        0.009
  inflation (-1)       (0.005)      (0.005)      (0.006)      (0.006)
Lagged GDP              -0.122       -0.130       -0.116       -0.114
  growth (-1)          (0.080)      (0.100)      (0.081)      (0.082)
Exchange rate         0.024 ***    0.026 ***    0.024 ***    0.024 ***
  regime               (0.004)      (0.004)      (0.004)      (0.004)
Government debt         0.052        0.021        0.041        0.039
                       (0.031)      (0.036)      (0.037)      (0.037)
Natural resources     0.079 ***     0.074 **    0.080 ***    0.079 ***
                       (0.029)      (0.031)      (0.027)      (0.028)
Ethnic                -0.043 **    -0.040 **    -0.041 **    -0.040 **
  polarization         (0.017)      (0.019)      (0.016)      (0.016)
Shadow economy 1      0.078 ***    0.108 ***    0.085 ***    0.084 ***
                       (0.027)      (0.035)      (0.030)      (0.030)
Lagged govt                          -0.001
  turnover (-1)                     (0.011)
Military                            -0.018 *
                                    (0.010)
Lagged constl                                   -0.028 ***   -0.029 ***
  change (-1)                                    (0.009)      (0.010)
Lagged cabinet                                    -0.005
  change (-1)                                    (0.006)
Lagged exec                                                   -0.0004
  change (-1)                                                 (0.007)
CBA

M2

1990-1991 dummy         0.018        0.014        0.019        0.019
                       (0.012)      (0.012)      (0.013)      (0.012)
Quadratic time        -0.431 ***   -0.404 **    -0.455 ***   -0.459 ***
  trend                (0.124)      (0.159)      (0.125)      (0.127)
Constant               -0.064 *    -0.092 **    -0.064 **    -0.066 **
                       (0.033)      (0.038)      (0.031)      (0.031)
F-test                11.61 ***    14.26 ***    10.37 ***    10.69 ***
Arellano-Bond test      0.000        0.000        0.000        0.000
  AR(1), p value
Arellano-Bond test      0.931        0.797        0.920        0.923
  AR(2), p value
Hansen test of          0.369        0.465        0.355        0.362
  overid., p value
Difference-in-          0.312        0.536        0.354        0.362
  Hansen tests,
  p value
Number of                 23           25           25           25
  instruments
Number of                123          123          121          121
  countries
Average obs. per        18.44        15.40        18.23        18.23
  country
Total panel obs.        2,268        1,894        2,206        2,206

                         (5)          (6)          (7)

Lagged RM 1 (-1)      0.210 ***    0.179 ***    0.153 ***
                       (0.055)      (0.060)      (0.048)
Lagged                0.016 ***    0.016 ***     0.009 *
  inflation (-1)       (0.005)      (0.004)      (0.005)
Lagged GDP              -0.187       -0.146     -0.164 **
  growth (-1)          (0.142)      (0.101)      (0.079)
Exchange rate         0.024 ***    0.028 ***    0.025 ***
  regime               (0.005)      (0.004)      (0.004)
Government debt         0.015        0.009        0.031
                       (0.037)      (0.039)      (0.039)
Natural resources       0.067       0.060 *     0.071 ***
                       (0.046)      (0.031)      (0.027)
Ethnic                -0.054 **    -0.041 **    -0.046 ***
  polarization         (0.021)      (0.020)      (0.017)
Shadow economy 1      0.137 ***     0.104 **     0.094 **
                       (0.042)      (0.050)      (0.040)
Lagged govt             0.006        0.0001
  turnover (-1)        (0.013)      (0.012)
Military               -0.026 *      -0.017
                       (0.013)      (0.011)
Lagged constl                                   -0.024 **
  change (-1)                                    (0.009)
Lagged cabinet                                    -0.002
  change (-1)                                    (0.006)
Lagged exec
  change (-1)
CBA                     0.022
                       (0.044)
M2                                   0.045        0.047
                                    (0.038)      (0.031)
1990-1991 dummy         0.017        0.013        0.019
                       (0.015)      (0.013)      (0.013)
Quadratic time         -0.309 *    -0.520 ***   -0.533 ***
  trend                (0.178)      (0.158)      (0.124)
Constant              -0.125 **    -0.101 ***   -0.070 **
                       (0.052)      (0.030)      (0.029)
F-test                16.57 ***    16.59 ***    13.13 ***
Arellano-Bond test      0.000        0.000        0.000
  AR(1), p value
Arellano-Bond test      0.625        0.800        0.542
  AR(2), p value
Hansen test of          0.526        0.342        0.443
  overid., p value
Difference-in-          0.586        0.416        0.368
  Hansen tests,
  p value
Number of                 26           26           26
  instruments
Number of                106          120          118
  countries
Average obs. per        14.70        15.09        17.92
  country
Total panel obs.        1,558        1.811        2,114

Notes: Windmeijer-corrected cluster-robust errors are in
parentheses. The null hypothesis in the Arellano-Bond test
for second-order serial correlation is that there is no
second-order serial correlation in the disturbances, in
differences. The null hypothesis in the Hansen test of over-
identified restrictions is that all instruments are valid.
The null hypothesis in the difference-in-Hansen tests of
exogeneity of the full set of GMM instruments for the levels
equation is that instruments are exogenous. The dependent
variable (Seigniorage 1) is measured as the change in
reserve money, divided by government spending. RM1 is
reserve money (line 14a of IMF-IFS) as percentage of
government spending.

* p <.1; ** p <.05; *** p <.01.