Do remittances induce inflation? Fresh evidence from developing countries.

Narayan, Paresh Kumar ; Narayam, Seema ; Mishra, Sagarika 等

1. Introduction

There is now a growing interest in the economic and institutional determinants of inflation, mainly because they have implications for monetary policy. As a result, there is a large and growing literature on the determinants of inflation (see, inter alia, Catao and Terrones 2005). There are only two studies in this regard that are closest to our work, but none of them have examined the impact of remittances on the inflation rate. Aisen and Veiga (2006) examined the relationship between inflation, political instability, and institutions using a data set covering 97 countries. Of these, 75 were developing countries. Their work was based on data covering the period ranging from 1960 to 1999. They used a dynamic panel data system generalized method of moments (GMM) estimator and within-group (fixed effects) estimator and found that a higher degree of political instability generates higher inflation. They also found that trade openness and economic growth reduced inflation, while growth in oil prices and the U.S. Treasury bill rate increased inflation.

Cottarelli, Griffiths, and Moghadam (1998) examined the determinants of inflation for a sample of 47 industrial and transition economies over the 1993-1996 period. They used an instrumental variables technique and found that fiscal deficits, trade openness, countries with a wage indexation mechanism, and an independent central bank all had a positive effect on inflation.

The goal of this article is to extend the work on the determinants of inflation for developing countries. We use a panel data set consisting of 54 developing countries over the 1995-2004 period. Our estimation technique is based on the Arellano and Bond (1991) panel dynamic estimator and the Arellano and Bover (1995) and the Blundell and Bond (1998) system GMM estimator.

We advance the literature forward in two novel ways. First, over the last couple of decades remittance to developing countries has increased steadily. We develop the link between inflation and remittances and empirically test whether remittances induce inflation. Remittances have emerged as a crucial issue for fiscal and monetary policy. Workers' remittances to developing countries have increased by a more than fivefold measure over the 1990-2004 period, from USD31.2 billion to USD160.4 billion, representing about 1.8% of developing countries' GDP (World Bank 2006). For many developing countries, remittances constitute the single largest source of foreign exchange. They have surpassed export revenues, foreign direct investment, and other private capital inflows (Amuedo-dorantes and Pozo 2004). The World Bank (2005) reports that while workers' remittances have increased from USD58 billion in 1995 to USD160 billion in 2004, over the corresponding period foreign direct investment increased from USD107 billion to USD166 billion, and official development assistance increased from USD59 billion to USD79 billion. The advent of remittances as a source of external finance is perceived as an avenue for fostering development and rescuing developing countries from both man-made and natural crises (see IMF 2005). Second, we model the determinants of inflation by taking into account political stability and institutional variables, namely government stability, military in politics, law and order, and democratic accountability. (1)

The balance of this article is organized as follows. In the next section, we discuss the theoretical framework. In section 3, we explain the model and the estimation technique. In section 4, we discuss the results. In the final section, we provide some concluding remarks.

2. Theoretical Underpinnings

Inflation and Remittance

The effect of remittances on inflation can be viewed from three different perspectives: namely, from the points of view of appreciating exchange rates, increasing money supply, and balance of payments. The rising level of remittances in developing economies can have a spending effect. This can trigger a rise in the price level of non-tradables. The Salter-Swan-Corden-Dornbusch paradigm offers an avenue for understanding the theoretical relationship among capital inflows (in our case, remittances), the price level, and the real exchange rate in developing economies. The model shows that an increase in remittances could cause a real exchange rate appreciation via rising domestic prices. More importantly, the extent of the effect of a rising level of remittances on domestic prices will depend on the country's exchange rate regime.

Reinhart and Rogoff (2004) show that different exchange rate regimes have considerably different effects on macroeconomic variables. Under a fixed exchange rate regime, for instance, an increase in remittances will move resources from the tradable to the non-tradable sector. This will result in an increase in the price level. Since the exchange rate is fixed, the country cannot adjust its international relative prices after a negative shock to the tradable sector. The nominal depreciation is, thus, prevented, and as a result, the tradable output contracts and the price level rises.

On the other hand, under a flexible exchange rate regime, since international relative prices can be adjusted following a large inflow of remittance, the resulting effect will be a rising price level and appreciation of the exchange rate. Rodrik (2007) provides evidence that the overvaluation of the real exchange rate (following an increase in remittances) causes an underestimation of long-term economic growth, particularly for developing economies. For these countries, the production of tradable goods suffers from weak institutions and market failures. This can potentially lead to an increase in inflation.

Acosta, Mandelman, and Lartey (2007) develop a micro-founded dynamic stochastic general equilibrium model that can explain the increasing price level when remittance is high. They consider a transmission mechanism: an increase in the household income (due to remittances) results in a decrease in the labor supply. A shrinking labor supply is associated with higher wages in terms of the price of the tradable output. This in turn leads to higher production costs, contributing to a further contraction of the tradable sector. Both the real exchange rate and the ratio of tradable to non-tradable output therefore induce high spending and resource movement. This can potentially result in an increase in inflation. Obstfeld and Rogoff (1996) contend that a positive transfer of resources to a country erodes its competitiveness in the global market because transfers lead to an appreciation of the real exchange rate. This implies that resource transfers generate inflation.

The increase in domestic prices due to a high exchange rate (which results from high remittances) can also be explained from the viewpoint of exchange rate pass-through; for a detailed discussion on this, see Choudhri and Hakura (2006).

When large inflows of foreign exchange are remitted by expatriates to their home country, the conversion of this foreign exchange into domestic currency raises the money supply. If this is not absorbed into productive sectors (or capital investment), then it goes into consumption expenditure, fuelling inflation. Remittances also boost real wealth, which stimulates consumption expenditure. This creates short-run excess demand, which drives up the price level.

The relationship between remittances and inflation can also be looked at from the point of view of the balance of payments and foreign reserves accumulation, as follows. Remittances can also be a source of balance-of-payments surplus and international reserves accumulation. Failure of central banks to fully sterilize the increase in international reserves will lead to an increase in the monetary base. This will result in further appreciation of the exchange rate. As a result, there will be an upward pressure on prices. Bugamelli and Paterno (2009) proposed a similar idea.

Inflation and Openness

The link between inflation and openness is well captured in the Barro and Gordon (1983) model, in which an unanticipated monetary expansion can cause inflation. The incentive to create surprise inflation is present if the rate of targeted output is lower than the natural rate of output. The central bank may target a rate higher than the natural rate if the natural rate is below the social optimum. Romer (1993) argued against this. He suggested that openness actually can reduce inflation since greater openness exposes a country, so much so that it suffers more from a real depreciation. In this case, there is less incentive for a country to pursue monetary expansion in open economies. Romer (1993) provides empirical evidence of this negative relationship between inflation and openness for a cross-section of countries. His findings are supported by Lane (1997).

Moreover, in the political science literature the "compensation hypothesis" is posited to explain a state's behavior in response to globalization. In the face of greater exposure to the outside world, the welfare state is seen as offsetting the social costs of international integration. The role of the state is crucial because despite the net economic gains from openness, developing countries exposed to international markets are likely to experience social dislocation, uncertainty, and unequal distributive effects. A related argument is provided by Rodrik (1997). Kaufman and Segura-Ubiergo (2001) argue that this can be a source of backlash against market-oriented economic policies, resulting in political and social instability. Government responds to cushion these anti-market oriented policies through the provision of welfare transfers. This assists people in coping with the structural change brought about by greater openness. By boosting spending, the development of the welfare state is likely to generate inflationary pressures.

Inflation and Institutions

The role of central bank independence has been a major issue in the literature on the determinants of inflation. Whether or not a central bank is independent depends in large part on whether a country is democratic. In most non-democratic countries, central banks are at the mercy of the government in that the central bank is open to political intervention, thus undermining the optimality of monetary policy. This is particularly the case during election time, when an incumbent government can use expansionary monetary policy to boost growth at the expense of high inflation. The International Monetary Fund (IMF 1996) found that between 1975 and 1995 the inflation performance of industrialized countries was negatively correlated with central bank independence. Other studies support this correlation (see Guitierrez 2003).

Some studies have found that democracies have better economic performances. The poor economic performances of non-democratic economies, characterized by unstable government, involvement of military in politics, and an absence of the rule of law, among other characteristics, is attributed to the predatory behavior of state officials. Non-democratic leaders use the state apparatus to extract revenue. They maintain themselves in power with repression and by using public spending and employment to create patronage networks and to co-opt popular unrest (Fatton 1992). The result, as clearly explained by Gasiorowski (2000, p. 321), is high inflation and slow growth, resulting from low investment levels, large fiscal deficits, loose monetary policy, chronic trade deficits, and inefficient state bureaucracies and public enterprises.

Rogoff (1985) argues that independent central banks ensure low inflation through solving the "time inconsistency problem," whereby governments attempt to stimulate the economy during election time. This action is aimed at stimulating growth and fighting unemployment--two appealing conditions for the electorate.

In democracies, party politics also have implications for inflation. Several authors have argued that the partisanship of the government does matter with regard to the type of macroeconomic policies adopted (see Hibbs 1977). The consensus is that left-wing governments are better at promoting economic growth than are right-wing governments (Whiteley 1983), while right-wing governments produce relatively smaller deficits than do left-wing governments (Roubini and Sachs 1989). In other words, left-wing governments focus on promoting economic growth and boosting employment--a constituency obligation, while right-wing governments focus on fighting inflation to protect the assets of the wealthy and the middle class--their target constituency (Kaltenthaler and Anderson 2000).

Inflation, Government Debt, and Current Account

There is a direct link between debt and monetary policy as a result of seigniorage. When an economy relies heavily on seigniorage revenues, its average level of seigniorage should be high. Thus, money financing is important. Sargent (1982) argues that European inter-war hyperinflation grew out of persistently active fiscal policies, which forced monetary authorities to adjust the money stock passively to meet high levels of government deficits.

A high level of debt could lead to monetization of debt, in turn leading to high inflation. The monetary policy implications of high debt when the real interest rate exceeds economic growth are well explained by Sargent and Wallace (1981). To achieve a lower interest rate, controlling inflation through reducing money growth would raise the debt:GDP ratio. The result will be an increase in interest payments and higher budget deficits. An expanding budget deficit will generate growth in money and inflation in the long run.

The fiscal theory of price determination, the proponents of which include Leeper (1989), Sims (1994), and Woodford (2001), among others, argues that if the real primary surplus is determined exogenously, the quantity of debt and price level may be related through a wealth effect. A rise in wealth through a non-Ricardian tax cut, for instance, stimulates consumption. The resulting excess demand generates a rise in price levels until excess demand falls and restores equilibrium. The fall in excess demand results from a decline in real wealth, which consists of real debt holdings. The maturity of government debt has direct implications for inflation. The greater the maturity, the smaller the inflation required to restore equilibrium following an expansionary fiscal policy.

Other Determinants of Inflation

The share of agriculture in GDP is used to model inflation. The agricultural share in GDP proxies a structural shift in the economy, and it captures a country's speed of transition and any setbacks due to financial crises when the agricultural sector acts as a shock absorber. A large agricultural sector relative to GDP implies a large resource transfer among sectors in the economy. Easy credit from the central bank ensures smooth adjustments, leading to a reduced impact on unemployment but higher inflation. Cukierman (1992) treated this as the employment motive for inflationary policies.

Cukierman, Edwards, and Tabellini (1992) consider the role of the agricultural sector as a revenue motive for inflationary policies. Their argument centers on the fact that the agricultural sector is the hardest to tax. This implies that countries in which the share of agriculture in GDP is high, the dependence on seigniorage revenue is greater. Higher seigniorage revenues lead to higher inflation.

Past inflation (lagged inflation) is also modeled as a determinant of inflation. Lagged inflation affects current inflation in two ways: (i) if past inflation has been costly, current inflation maybe lower; and (ii) if inflation expectations are backward looking, a relatively high past inflation would make disinflation more costly, thereby resulting in higher inflation equilibrium (Cottarelli, Griffiths, and Moghadam 1998).

3. Model Specification

Based on the conceptual framework presented in the previous section, which also provides the theoretical motivation for including lagged levels of the inflation rate as an additional regressor, we have a dynamic panel specification. Our preference, given the relatively small panel that we have, is to use the Arellano and Bond (AB, 1991) GMM estimator and the Arrellano and Bover (1995) and the Blundell and Bond (1998) system GMM estimator. Our proposed model takes the following form:

[inf.sub.i,t] = [alpha]0[inf.sub.i,t] - 1 + [beta][X'.sub.i,t] + [PSI][gremit'.sub.i,t] + [v.sub.i] + [[epsilon].sub.i,t], i =1,..., N; t = 1,..., T (1)

where inf stands for the inflation rate of country i at time t; gremit stands for the growth rate in remittances as a percentage of GDP (henceforth remittance growth); X is a vector of core explanatory variables used to model inflation, apart from remittances; v is country-specific effects; and [epsilon] is the error term. The model was also estimated by using log[l + (inf/100)] as a dependent variable.

By comparison, using the pooled ordinary least squares (OLS) and the panel OLS estimator (with fixed and random effects) is problematic because the lagged dependent variable is correlated with the error term. By first differencing Equation l, the AB GMM estimator solves this problem and eliminates country-specific effects. E([[epsilon].sub.i,t] - [[epsilon].sub.i,t-1]) = 0, but ([inf.sub.i,t] -[inf.sub.i,t - 1]) is not independent of ([[epsilon].sub.i,t] - [[epsilon].sub.i,t-1]). The AB method solves this problem by using two or more lags of the first difference of inflation as instruments. We use two lags. With respect to ([X.sub.i,t] - [X.sub.i,t -1]) and ([gremit.sub.i,t] - [gremit.sub.i,t - 1]), we assume that all the control variables except output growth, remittance growth, and government stability are predetermined in the sense that E([X.sub.i,t], [[epsilon].sub.i,s]) [not equal to 0] and for s < t but 0 for s > [greater than or equal to] t. For the predetermined variables, one or more period lagged levels of the variables are orthogonal to the differenced error term and thus form valid instruments for respective first differenced right-hand side variables. We use two lags.

We allow for possible endogeneity between inflation and government stability, inflation and remittance growth, and inflation and output growth because current period inflation could potentially affect the flow of remittance and the output growth rate. The higher the inflation, the higher could be the flow of remittance and the lower could be the output growth. Similarly, current inflation can affect the stability of the political party in power. The government will become more unstable as inflation increases. To account for these endogeneity issues, we include the one period lagged value of these variables as a valid instrument in the regression. Our sample size is close to 500 observations, and the Sargan test results support a two-step estimator over the one-step estimator. It follows that in our empirical work we use the two-step estimator.

4. Data and Results

Data

Our empirical analysis is based on annual data covering the 1995-2004 period for 54 developing countries. These countries include the following: Albania, Argentina, Armenia, Azerbaijan, Bangladesh, Bolivia, Botswana, Brazil, Cameroon, China, Colombia, Congo Republic, Costa Rica, the Dominican Republic, Ecuador, Egypt, El Salvador, Ethiopia, Gabon, Ghana, Guatemala, Guinea, Guyana, Haiti, Honduras, India, Indonesia, Jamaica, Jordon, Latvia, Lithuania, Malawi, Mali, Mexico, Moldova, Morocco, Namibia, Nicaragua, Niger, Nigeria, Pakistan, Panama, Paraguay, Peru, Philippines, Senegal, Sierra Leone, Slovenia, Sri Lanka, Sudan, Togo, Trinidad and Tobago, Tunisia, and Yemen. In all, our sample of countries includes 19 African countries, 17 Central and South American countries, eight European countries, and seven Asian countries. (2)

Data are obtained from four sources: the World Development Indicators (WDI), the International Financial Statistics (IFS), Bloomberg, and the International Country Risk Guide (ICRG). Data on inflation, real economic growth rate, agricultural sector output as a percentage of GDP, trade openness (sum of exports plus imports as a percentage of GDP), current account deficit as a percentage of GDP, total debt as a percentage of GDP, nominal exchange rate, and workers' remittances are from the WDI. We use the growth rate of remittances as a percentage of GDP. Data on the U.S. short-term Treasury bill rate are from the IFS. Data on crude oil prices are from Bloomberg. Data on democratic accountability, government stability, law and order, and the military in politics are obtained from the ICRG. The government stability index measures the government's ability to carry out its policies and its ability to stay in office for the full duration. It is made up of three components: government unity, legislative strength, and popular support. Points range from 0 to 12, with 12 indicating the highest level of government stability. The military in politics index measures the involvement of the military in politics. Points range from 0 to 6, with 6 indicating the lowest level of military participating in politics.

The law and order index measures the strength and impartiality of the legal system and the popular observance of the law, with points ranging from 0 to 6. Six implies the highest level of law and order. The democratic accountability index measures the responsiveness of government to its people. The less responsive it is, the more likely it is that the government will fall, peacefully in a democratic society but possibly violently in a non-democratic one. Points range from 0 to 6, with 6 indicating the highest level of democratic accountability.

Main Findings

We start our analysis by checking for outliers in the data. We apply the "mean [+ or -] 3 standard deviations (SD)" rule to remove outliers from each of the variables for each country. However, we did not find any outliers. In Table 1, we report descriptive statistics of our variables. The mean inflation for our panel is 9.7%, and the SD is 11.1%. This indicates that the variability of inflation is relatively high across countries in our panel. This is expected because we are only considering developing countries for our analysis. The mean current account deficit as a percentage of GDP is very low, indicating that the panel has a higher change in net foreign asset with a lower variability. This is because these countries have high inflow of remittances that cause a current account reversal (Bugamelli and Paterno 2009). The statistics for all the institutional variables are very similar across the panel.

We start our empirical analysis by running pooled OLS and fixed effect estimations. Although the coefficient estimates are inconsistent in both the cases, we run these regressions to compare the coefficients with the GMM estimation. The OLS and fixed effect estimation results are shown in Table 2. The results indicate that lag inflation is positive and significant and output growth is negative and significant under both estimators. However, remittance is not statistically significant in either of the two models. Log nominal exchange rate is negative and significant, consistent with our expectations. Current account deficit is insignificant. Total debt is positive and significant. Trade is positive and significant under fixed effect estimation but negative and significant under OLS.

Next, we estimate system and differenced GMM models for our panel. The system and differenced results are shown in Tables 3 and 4, respectively. We estimate several variants of the inflation model to test the robustness of the impact of remittances on the inflation rate for 54 developing countries. In model 1, we examine the impact of the economic growth rate, the nominal exchange rate, and the remittance growth rate on the inflation rate. In addition to the above three variables, in models 2-4 we examine the impact of trade openness, the current account deficit, and total debt on the inflation rate. In addition to all of the above explanatory variables, in models 5-7 we successively add agricultural output, growth of crude oil prices, the U.S. interest rate, democracy, government stability, military, and law and order, respectively.

System GMM

We find that for all specifications, lag inflation is positive and significant. Consistent with economic theory, output growth is negative and significant in models 7, 8, and 10. (3) Remittance growth rate has a statistically significant and positive effect on the inflation rate in 10 out of 11 models. This indicates that high levels of remittances induce inflation. Consistent with theory, the nominal exchange rate has a statistically significant negative effect on inflation in 9 out of 11 models. Trade has a positive and statistically significant coefficient in only 2 out of 10 models. The current account deficit as a percentage of GDP has a statistically significant positive effect on the inflation rate in only four out of nine models. Total debt as a percentage of GDP has a positive and statistically significant effect in all eight models. Agricultural output as a percentage of GDP and the U.S. interest rate have a positive and significant effect on the inflation rate, whereas the growth rate of crude oil price does not have any effect on inflation. Among the four institutional variables, democracy and military in politics have statistically significant negative effects on inflation, whereas government stability and law and order do not have any significant effects on inflation.

The results from models 1-11 reveal two important features. First, we find that the inclusion of additional regressors does not change the results on the impact of remittances on the inflation rate. For instance, across all 10 models, we find the magnitude of the impact of the growth rate of remittances to be in the range of 0.001 to 0.009. Second, we notice that the magnitude and statistical significance of the rest of the variables are fairly consistent across all models.

Differenced GMM

For all specifications, lag inflation is found to be positive and statistically significant. Consistent with economic theory, the economic growth rate has a statistically significant negative effect in all models. Remittance growth rate has a statistically significant positive effect on the inflation rate in 3 out of 11 models. Nominal exchange rate has a negative and statistically significant effect on inflation in all 11 models. Trade has a positive and significant coefficient in all 10 models. The current account deficit as a percentage of GDP has a statistically significant positive effect on the inflation rate in all nine models. Total debt as a percentage of GDP has a positive and significant effect in all eight models. Agricultural output as percentage of GDP and the U.S. interest rate have a positive and significant effect on the inflation rate, whereas the growth rate of crude oil price does not have any significant effect on inflation. Among the four institutional variables, democracy, government stability, and military in politics have statistically significant negative effects on inflation. Meanwhile, law and order does not have any significant effect on inflation. Although in the short run (results from differenced GMM estimation) the addition of other regressors makes the remittance growth rate insignificant, in the long run (results from system GMM estimation) growth of remittance has a significantly positive effect on inflation. The sign and significance of all other variables mostly remain consistent across all models.

To give credence to our results, we run some diagnostic tests to examine whether the data are consistent with the assumptions of the AB estimator. In particular, we report the Sargan test statistic, which examines the over-identification restrictions. It essentially tests whether the instruments are uncorrelated with the error terms in the estimated equation. The null hypothesis is that the instruments as a group are exogenous. A finding of exogenous instruments is needed for the validity of the GMM estimates. The Sargan test statistic (together with its associated p-values) is reported in the second last two rows of Tables 3 and 4. The Sargan test statistics for all models appear with p-values greater than 0.10; hence, we are unable to reject the null hypothesis.

The second test we report is the Arellano and Bond test for autocorrelation. The null hypothesis is "no autocorrelation" and relates to the differenced residuals. We only report the test statistics and their associated p-values for the autoregressive model with two lags because it detects autocorrelation in levels. For both system GMM and differenced estimation for all of the 11 models we are unable to reject the null hypothesis of "no autocorrelation." There is robust evidence that for all 11 models there is no autocorrelation at the 1% level.

Discussion of Results

Based on the pooled OLS and the fixed effect estimations, we did not find remittance growth to be a statistically significant determinant of inflation. This is not surprising because the coefficient estimates are biased under the OLS and the fixed effect estimators. Based on the system GMM estimator, we find that across 11 different models of the determinants of the inflation rate, remittances had a statistically significant positive effect on the inflation rate in developing countries. This indicates that remittances exert inflationary pressures in developing countries in the long run. In the short run, we also find some evidence, although not as conclusive as in the case of the long run, that remittances induce inflation. The findings indicate that remittances are not absorbed into productive sectors (or capital investment); rather, they go toward consumption expenditure. This fuels inflation. This behavior of remittances in developing countries is also consistent with the wealth transfer inflationary pressure argument developed by Obstfeld and Rogoff (1996). The main thesis of their argument is that resource transfers deplete a country's global market competitiveness, triggering an appreciation of the real exchange rate, thus further fueling inflation.

With altruistically motivated remittances, between remitters and their family members exists a situation of asymmetric information whereby remitters are incapable of monitoring how the remitted funds are used. If used for capital investment, remittances offer a means of building a sustainable livelihood. However, in most cases, remittances are used for consumption purposes. This is noted in the work of Meyers (1998).

Consistent with economic theory, in all models we find that the nominal exchange rate had a statistically significant negative effect on inflation. This is because when the exchange rate appreciates it reduces net exports by making imports cheaper. The reduction in net exports shifts the aggregate demand curve leftward. For a given aggregate supply, this will result in a reduction in the price level.

Our result shows that openness has a statistically significant positive effect on inflation in most of our models. This result is consistent with the compensation hypothesis, in which greater openness leads to unequal distribution of gains from openness. In other words, openness exposes developing countries to greater social and economic inequality, generating political demand for social spending, aimed at providing compensation for increased risk associated with openness. To avert political and economic crises, governments respond by developing a welfare state, capable of ensuring parity in the distribution of gains from greater trade. The emergence of the welfare state stimulates consumer spending, thereby generating inflationary pressures.

A second explanation for our results has roots in the work of Kydland and Prescott (1977), who show that when central banks are not pre-committed to monetary policy, there is high inflation. The main tenet of their idea is based on imperfect competition arising from trade openness. If this results in suboptimal output levels, and when monetary policy can stimulate output to its natural level, the central bank has an incentive to create surprise inflation.

While consistent with the results of the Kydland and Prescott (1977) hypothesis and the compensation hypothesis, our results on openness are inconsistent with the literature that has used similar panel data estimation approaches to our work. For example, Desai, Olofsgard, and Yousef (2003) examined the relationship between inflation and openness for a panel of 100 countries (both developed and developing) and found that trade openness reduces inflation. Aisen and Veiga (2006) examine the inflation-openness nexus for a panel of 75 developing countries and find an inverse relationship between openness and inflation. In a recent study, Bowdler and Nunziata (2006), for a panel of 18 OECD countries, found that increased trade openness reduces the probability of inflation. Similar results were found by Romer (1993), who argued that one should observe a negative openness-inflation relationship, if such a mechanism is a driving force of inflation.

We find a fairly consistent negative and statistically significant relationship between economic growth and inflation. An increase in economic growth implies that there are more goods and services for a fixed amount of money, an increase in goods and services relative to money. This puts downward pressure on inflation. This finding is consistent with the literature that has modeled the impact of economic growth on inflation for panels of countries. For example, Aisen and Viega (2006) found a negative relationship between economic growth and inflation for a panel of 75 developing countries, and Desai, Olofsgard, and Yousef (2003) found a negative relationship between economic growth and inflation for a panel of 100 developed and developing countries.

Our results on the relationship between institutions and the inflation rate in developing countries are mixed in terms of statistical significance. For instance, in the short run, an improvement in democracy, government stability, and military in politics has a statistically significant negative effect on the inflation rate, but law and order has a statistically insignificant effect on the inflation rate. In the long run, government stability does not have any significant effect on inflation. However, democracy and military in politics have a negative effect on inflation in the long run.

Our results on the impact of democracy are consistent with similarly motivated studies. For example, Aisen and Veiga (2006), for a panel of 75 developing countries, found that political instability, increased government crises, and cabinet changes all contributed to higher inflation in developing countries. Consistent with our findings, they report that democracy reduces inflation. However, our findings and those of Aisen and Viega are inconsistent with those of Gasiorowski (2000), who found that democracy increased the inflation rate in a panel of 49 developing countries.

We find that our inclusion of variables capturing external conditions, namely the shortterm U.S. Treasury bill rate, creates inflationary pressures. This finding implies that external conditions, in particular the U.S. monetary policy, is a source of inflation in developing countries. Developing countries are dependent on imported goods (capital) for their production. A hike in foreign interest rates makes the financing of imported capital more expensive. This adversely affects aggregate supply and exerts inflationary pressures. In this literature, one previous study has modeled the impact of the U.S. interest rate and oil prices on developing country inflation rates: Aisen and Veiga (2006), for a panel of 75 developing countries, found that the U.S. interest rate and oil prices exert inflationary pressures. However, we did not find oil prices to have a significant effect on inflation.

We find that current account deficits and total debt (both measured as percentage of GDP) exert inflationary pressures in developing countries. Deficits and debts generate inflationary pressures through the monetization of deficit/debt, promoting central banks to ensure that the government's intertemporal budget is balanced through increasing seigniorage (see Sargent and Wallace 1981). Thus, there is a reliance on inflation to finance government expenditure in developing countries. In comparison with the findings from the literature, Cottarelli, Griffiths, and Moghadam (1998), for instance, found a positive but statistically insignificant effect of the current account deficit on inflation and a statistically significant positive effect of the fiscal deficit on inflation. In a related study, Gasiorowski (2000), for a panel of developing countries, found that fiscal deficits had a statistically significant positive effect on inflation.

Our results also indicate that a larger share of the agricultural sector output in GDP generates inflation. This observation was first made by Cukierman, Edwards, and Tabellini (1992), who argued that the agricultural sector is the hardest to tax. This implies that in countries in which the share of agriculture in GDP is high, the dependence on seigniorage revenue is greater. Using a cross-section of 79 countries (both developing and developed), Cukierman, Edwards, and Tabellini (1992) show that a growing agricultural sector has a statistically significant positive effect on seigniorage. Higher seigniorage revenues lead to higher inflation.

Robustness Check

To check robustness of our results, we use log[1 + (inf/100)] instead of inflation as the dependent variable to estimate the models. The results are shown in Table 5 (system GMM) and Table 6 (differenced GMM). Based on the system GMM results, we find that remittance growth rate has a positive and significant effect on inflation in most cases. All other variables have coefficients of similar sign and significance, as reported previously. Based on results obtained from the differenced GMM estimation, we find that remittance is statistically insignificant in most of the models. This implies that while in the short run remittance is mostly insignificant, over the long run remittance does significantly induce inflation in developing countries. We also exclude the Latin American countries from the sample and re-run the regression, and the results do not change much. In subsequent subsample analysis, we also exclude Asian countries and reestimate these models, but no significant change in results is found. For all of the above estimations, the models pass the Sargan and second-order autocorrelation tests.

5. Concluding Remarks

There is a growing interest in panel data analysis of the determinants of inflation. In this study, we take this literature forward through modeling the impact of remittances on the inflation rate in a panel of 54 developing countries over the period ranging from 1995 to 2004. Given the small panel and the commonly acknowledged problems with panel fixed effects estimators, we use the AB dynamic panel estimator and the Arellano and Bover (1995) and the Blundell and Bond (1998) system GMM estimator that corrects for endogeneity of regressors through use of appropriate instruments. Because we modeled the impact of remittances on inflation, and given the originality of the empirical analysis, an important aspect of our analysis was to confirm the robustness of the impact of remittances on the inflation rate.

After estimating as many as 11 different models of the determinants of inflation, we found fairly consistent evidence that remittances across the bulk of the models had a positive and statistically significant effect on the inflation rate. This implies that remittances generate inflationary pressures in developing countries. Among our other main results, we find that (i) improvements in democracy and the involvement of military in politics reduce inflation rates; (ii) current account deficits, debts, the U.S. interest rate, and the agricultural sector output increase inflation rates: (iii) openness generates a rise in inflation; and (iv) economic growth rate decreases inflation rate, while the lagged inflation rate has a positive effect on current inflation.

The positive relationship between openness and inflation implies that central banks in developing countries have the incentive to create surprise inflation in order to boost suboptimal economic growth. There also appears to be public pressure on developing country governments to compensate those that suffer from a country's greater exposure to the international markets, leading to the creation of a welfare state.

References

Acosta, Pablo A., Federico S. Mandelman, and Emmanuel K. K. Lartey. 2007. Remittances, exchange rate regimes, and the Dutch disease: A panel data analysis. Federal Reserve Bank of Atlanta Working Paper.

Aisen, Ari, and Francisco Jose Veiga. 2006. Does political instability lead to higher inflation? A panel data analysis. Journal of Money, Credit and Banking 38: 1379-89.

Amuedo-dorantes, Catalina, and Susan Pozo. 2004. Workers' remittances and the real exchange rate: A paradox of gifts. World Development 32:1407 -17.

Arellano, Manuel, and Stephen Bond. 1991. Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Review of Economic Studies 58:277-97.

Arellano, Manuel, and Olympia Bover. 1995. Another look at the Instrumental Variable Estimation of ErrorComponent Models. Journal of Econometrics 68:29-51.

Barro, Robert J., and David B. Gordon. 1983. Rules, discretion and reputation in a model of monetary policy. Journal of Monetary Economics I2:101-21.

Blundell, Richard, and Stephen Bond. 1998. Initial Conditions and Moment Restrictions in Dynamic Panel Data models. Journal of Econometrics 87:115-43.

Bowdler, Christopher, and Luca Nunziata. 2006. Trade openness and inflation episodes in the OECD. Journal of Money, Credit and Banking 38:553-63.

Bugamelli, Matteo, and Francesco Paterno. 2009. Do workers' remittances reduce the probability of current account reversals? World Development 37:1821-38.

Catao, Luis A. V., and Marco E. Terrones. 2005. Fiscal deficits and inflation. Journal of Monetary Economics 52:529-54.

Choudhri, Ehsan U., and Dalia S. Hakura. 2006. Exchange rate pass through to domestic prices: Does the inflationary environment matter? Journal of International Money and Finance 25:614-39.

Cottarelli, Carlo, Mark Griffiths, and Reza Moghadam. 1998. The non-monetary determinants of inflation: A panel data study. International Monetary Fund Working Paper No. 23.

Cukierman, Alex. 1992. Central bank strategy, credibility, and independence: Theory, and evidence. Cambridge, MA: MIT Press.

Cukierman, Alex, Sebastian Edwards, and Guido Tabellini. 1992. Seignioriage and political instability. American Economic Review 82:537-55.

Desai, Raj M., Anders Olofsgard, and Tarik M. Yousef. 2003. Democracy, inequality and inflation. American Political Science Review 97:391-406.

Fatton, Robert, Jr. 1992. Predatory rule: State and civil society in Africa. Boulder, CO: Lynne Rienner.

Gasiorowski, Mark J. 2000. Democracy and macroeconomic performance in underdeveloped countries: An empirical analysis. Comparative Political Studies 33:319-49.

Ghosh, Atish, and Steven Phillips. 1998. Warning: Inflation may be harmful to your growth. Staff Papers--International Monetary Fund 45:672-710.

Guitierrez, Eva. 2003. Inflation performance and constitutional central bank independence: Evidence from Latin American and the Caribbean. International Monetary Fund Working Paper No. 53.

Hibbs, Douglas A., Jr. 1977. Political parties and macroeconomic policy. American Political Science Review 71:1467-87.

International Monetary Fund (IMF). 1996. The rise and fall of inflation: Lessons from the postwar experience. In World economic outlook. Washington, DC: IMF, pp. 100-242.

IMF (International Financial Statistics). 2005. World economic prospects: Globalisation and external imbalances. Washington, DC: IMF.

Jacome, Luis I. 2001. Legal central bank independence and inflation in Latin America during the 1990s. International Monetary Fund Working Paper No. 01/212.

Jacome, Luis I., and Francisco Vazquez. 2005. Any link between legal central bank independence and inflation? Evidence from Latin American and the Caribbean. International Monetary Fund Working Paper No. 05/75.

Kaltenthaler, Karl C., and Christopher J. Anderson. 2000. The changing political economy of inflation. Journal of Public Policy 20:109-31.

Kaufman, Robert R., and Alex Segura-Ubiergo. 2001. Globalization, domestic politics and social spending in Latin America: A: Time-series cross-section analysis 1973-1997. Word Politics 53:553-87.

Khan, Mohsin S., and Abdelhak S. Senhadji. 2001. Threshold effects in the relationship between inflation and growth. Staff Papers International Monetary Fund 48:1-21.

Kydland, Finn E., and Edward C. Prescott. 1977. Rules rather than discretion: The inconsistency of optimal plans. Journal of Political Economy 85:473-92.

Lane, Philip R. 1997. Inflation in open economies. Journal of International Economics 42:327-47.

Leeper, Eric M. 1989. Policy rules, information and fiscal effects in a Ricardian model. International Finance Discussion Paper 360. Federal Reserve Board.

Meyers, Deborah Waller. 1998. Migrant remittances to Latin America: Reviewing the literature. Unpublished manuscript: Tomas Rivera Policy Institute, University of Southern California.

Obstfeld, Maurice, and Kenneth S. Rogoff. 1996. Foundations of international macroeconomics. Cambridge, MA: MIT Press.

Reinhart, Carmen M., and Kenneth S. Rogoff. 2004. The modern history of exchange rate arrangements: A reinterpretation. Quarterly Journal of Economics 119:1-47.

Rodrik, Dani. 1997. Has globalisation gone too far? Washington, DC: Institute for International Economics.

Rodrik, Dani. 2007. The real exchange rate and economic growth: Theory and evidence. Unpublished manuscript, mimeo. Harvard University.

Rogoff, Kenneth. 1985. The optimal degree of commitment to an intermediate monetary target. Quarterly Journal of Economics 100:1169-90.

Romer, David. 1993. Openness and inflation: Theory and evidence. Quarterly Journal of Economics 108:869-903.

Roubini, Nouriel, and Jeffrey D. Sachs. 1989. Political and economic determinants of budget deficits in the industrial democracies. European Economic Review 33:903-38.

Sargent, Thomas J. 1982. The end of four big inflations. In Inflation." Causes and effects, edited by R. E. Hall. Chicago, IL: University of Chicago Press, pp. 41-97.

Sargent, Thomas J., and Neil Wallace. 1981. Some unpleasant monetary arithmetic. Quarterly Review, 5. Federal Reserve Bank of Minneapolis, pp. 1-17. Sims, Christopher A. 1994. A simple model for the study of the determination of the price level and the interaction of monetary and fiscal policies. Economic Theory 4:381-99.

Whiteley, Paul F. 1983. The political economy of economic growth. European Journal of Political Research 11 : 197-213.

Woodford, Michael D. 2001. Fiscal requirements for price stability. NBER Working Paper No. 8072.

World Bank. 2005. Global economic prospects. Washington, DC: World Bank.

World Bank. 2006. Global economic prospects: Economic implications of remittances and migration. Washington, DC: The International Bank for Reconstruction and Development, World Bank.

(1) While the literature has identified the importance of central bank independence in influencing the inflation rate (see, for instance, Jacome [2001] and Jacome and Vazquez [2005]), we do not model central bank independence here due to lack of data for the sample of countries considered in this article.

(2) At the suggestion of one referee of this journal, we also re-estimate our models by excluding Latin American countries and Asian countries successively. The results do not change much.

(3) Following Ghosh and Phillips (1998) and Khan and Scnhadji (2001), we account for the non-linearity relation between inflation and output growth by adding polynomials of output growth in our regressions. We do not find any of the coefficients to be statistically significant.

Paresh Kumar Narayan, Professor of Finance, School of Accounting, Economic's and Finance, 70 Edgar Road, Burwood, 3125, Melbourne, Australia: E-mail paresh.narayan@deakin.edu.au: corresponding author.

Seema Narayam, Senior Lecturer in Economics, School of Marketing, Economics and Finance, Royal Melbourne Institute of Technology, Melbourne, Australia: E-mail seema.narayan@rmit.edu.au.

Sagarika Mishra, Lecturer in Finance, School of Accounting, Economics and Finance, 70 Edgar Road. Burwood, 3125. Melbourne, Australia: E-mail sagarika.mishra@deakin.edn.au.

We would like to thank two anonymous referees of this journal for several invaluable comments and suggestions on earlier versions of this article. Errors or omissions, if any, are, however, our own doing.

Received July 2009; accepted June 2010.

Table 1. Descriptive Statistics

Variables                                 Mean       Median      SD

Inflation                                  9.72       6.67      11.14
Remittance growth                          7.32       5.18      76.58
Real economic growth rate                  3.66       4          2.81
Current account deficit as a % of GDP     -3.37      -3.36       6.96
Agricultural sector output as a % of      19.87      16.83      13.1
  GDP
Total debt as a % of GDP                  67.72      51.91      63.61
Trade openness measured as exports
  plus imports as a % of GDP              74.01      71.01      35.61
Short-term U.S. interest rate              3.92       4.84       1.79
Growth rate in crude oil prices            9         14.36      25.74
Log of nominal exchange rate               3.25       2.72       2.61
Democracy index                            3.66       4          1.32
Government stability index                 8.78       9.12       1.7
Military in politics index                 3.24       3          1.61
Law and order index                        3.33       3          1.12

SD indicates standard deviation.

Table 2. OLS and Fixed Effect Estimation Results

                                              OLS

Inflation.L1                           0.5573 *** (0.000)
GDP growth                            -0.9526 *** (0.000)
Remittance                                 0.0011 (0.754)
Log of nominal exchange rate          -0.3585 *** (0.001)
Current account deficit (% of GDP)         0.0512 (0.205)
Total debt (% of GDP)                  0.0197 *** (0.000)
Trade (% of GDP)                        -0.0156 * (0.052)
Intercept                              7.7692 *** (0.000)
No. of observations                           486
Adjusted [R.sup.2]                            0.61

                                          Fixed Effect

Inflation.L1                           0.3977 *** (0.000)
GDP growth                            -1.0614 *** (0.000)
Remittance                                 0.0001 (0.967)
Log of nominal exchange rate          -2.9729 *** (0.002)
Current account deficit (% of GDP)         0.0731 (0.202)
Total debt (% of GDP)                  0.0448 *** (0.002)
Trade (% of GDP)                       0.1236 *** (0.000)
Intercept                                  6.4543 (0.087)
No. of observations                           486
Adjusted [R.sup.2]                            0.21

*, *** denote statistical significance at the 10% and 1% levels,
respectively. The probability values are reported in  parentheses.

Table 3. System GMM Estimation Results

                                   Model 1           Model 2

Inflation.L1                     0.5019 ***        0.5087 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.0516            0.0882
                                (0.084)           (0.103)
Remittance growth                0.0095 ***        0.0060 ***
                                (0.000)           (0.000)
Log of nominal exchange rate     0.0220           -0.1807
                                (0.674)           (0.240)
Trade (% of GDP)                                   0.0333 ***
                                                  (0.000)
Current account deficit
  (% of GDP)
Total debt (% of GDP)

Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability

Military

Law and order

Sargan test                     52.20             44.72
                                (0.49)            (0.42)
No. autocorrelation             -1.16             -0.87
                                (0.25)            (0.38)
No. of observations            486               486

                                   Model 3           Model 4

Inflation.L1                     0.4783 ***        0.5079 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.0286           -0.0776
                                (0.641)           (0.240)
Remittance growth                0.0083 ***        0.0060 ***
                                (0.000)           (0.000)
Log of nominal exchange rate    -0.5655 ***       -0.4998 ***
                                (0.000)           (0.000)
Trade (% of GDP)                 0.0221 ***        0.0020
                                (0.000)           (0.857)
Current account deficit          0.0162            0.0279
  (% of GDP)                    (0.371)           (0.135)
Total debt (% of GDP)                              0.0181 ***
                                                  (0.000)
Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability

Military

Law and order

Sargan test                     43.96             40.27
                                (0.53)            (0.51)
No. autocorrelation             -1.02             -1.23
                                (0.31)            (0.22)
No. of observations            486               486

                                   Model 5           Model 6

Inflation.L1                     0.4791 ***        0.4921 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.0274            0.0093
                                (0.697)           (0.930)
Remittance growth                0.0025 ***        0.0055 ***
                                (0.000)           (0.000)
Log of nominal exchange rate    -0.3913 ***       -0.6129 ***
                                (0.000)           (0.000)
Trade (% of GDP)                 0.0196           -0.0079
                                (0.123)           (0.268)
Current account deficit          0.0520 *          0.0458 **
  (% of GDP)                    (0.057)           (0.003)
Total debt (% of GDP)            0.0134 ***        0.0199 ***
                                (0.000)           (0.000)
Agricultural output              0.1777 ***
  (% of GDP)                    (0.000)
Crude oil prices growth (%)                        0.0008
                                                  (0.604)
U.S. interest rate

Democracy

Government stability

Military

Law and order

Sargan test                     42.70             43.96
                                (0.28)            (0.44)
No. autocorrelation             -1.41             -1.33
                                (0.16)            (0.19)
No. of observations            486               486

                                   Model 7           Model 8

Inflation.L1                     0.5129 ***        0.4952 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.1101 *         -0.1295 *
                                (0.023)           (0.037)
Remittance growth                0.0065 ***        0.0042 ***
                                (0.000)           (0.000)
Log of nominal exchange rate    -0.5145 ***       -0.5456 ***
                                (0.000)           (0.000)
Trade (% of GDP)                 0.0022           -0.0117
                                (0.791)           (0.327)
Current account deficit          0.0407 **         0.0371 **
  (% of GDP)                    (0.013)           (0.003)
Total debt (% of GDP)            0.0165 ***        0.0196 ***
                                (0.000)           (0.000)
Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate               0.1094 ***
                                (0.001)
Democracy                                         -0.1835 *
                                                  (0.048)
Government stability

Military

Law and order

Sargan test                     38.02             37.91
                                (0.45)            (0.67)
No. autocorrelation             -1.36             -1.41
                                (0.18)            (0.16)
No. of observations            486               486

                                   Model 9          Model 10

Inflation.L1                     0.5116 ***        0.5002 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.0844           -0.1399 *
                                (0.146)           (0.080)
Remittance growth                0.0061 ***        0.0018
                                (0.000)           (0.122)
Log of nominal exchange rate    -0.5856 ***       -0.5464 ***
                                (0.000)           (0.000)
Trade (% of GDP)                -0.0096            0.0088
                                (0.268)           (0.445)
Current account deficit          0.0193            0.0075
  (% of GDP)                    (0.183)           (0.805)
Total debt (% of GDP)            0.0206 ***        0.0183 ***
                                (0.000)           (0.000)
Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability             0.0086
                                (0.917)
Military                                          -0.5766 *
                                                  (0.046)
Law and order

Sargan test                     36.06             44.28
                                (0.68)            (0.48)
No. autocorrelation             -1.34             -1.40
                                (0.18)            (0.16)
No. of observations            486               486

                                  Model 11

Inflation.L1                     0.4910 ***
                                (0.000)
GDP growth (%)                  -0.0285
                                (0.637)
Remittance growth                0.0042 ***
                                (0.001)
Log of nominal exchange rate    -0.5918 ***
                                (0.000)
Trade (% of GDP)                -0.0024
                                (0.857)
Current account deficit          0.0411
  (% of GDP)                    (0.150)
Total debt (% of GDP)            0.0220 ***
                                (0.000)
Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability

Military

Law and order                    0.2787
                                (0.378)
Sargan test                     49.79
                                (0.11)
No. autocorrelation             -1.25
                                (0.21)
No. of observations            486

p-values are in parentheses. *, **, and *** denote significance at
10%, 5%, and 1% levels of significance,  respectively. L1 stands for
one-period lag.

Table 4. Differenced GMM Results

                                   Model 1           Model 2

Inflation.L1                     0.3713 ***        0.4054 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.6817 ***       -0.5367 ***
                                (0.000)           (0.000)
Remittance growth                0.0069 ***        0.0037 ***
                                (0.001)           (0.000)
Log of nominal exchange rate    -5.0679 ***       -4.5978 ***
                                (0.000)           (0.000)
Trade (% of GDP)                                   0.1045 ***
                                                  (0.000)
Current account deficit
  (% of GDP)
Total debt (% of GDP)

Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability

Military

Law and order

Sargan test                     39.40             48.16
                                (0.45)            (0.70)
No. autocorrelation             -1.44             -1.26
                                (0.15)            (0.21)
No. of observations            432               432

                                   Model 3           Model 4

Inflation.L1                     0.3789 ***        0.3855 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.5076 ***       -0.3741 ***
                                (0.000)           (0.000)
Remittance growth                0.0034 ***        0.0020
                                (0.002)           (0.364)
Log of nominal exchange rate    -7.2046 ***       -4.8779 ***
                                (0.000)           (0.000)
Trade (% of GDP)                 0.1040 ***        0.0818 ***
                                (0.000)           (0.000)
Current account deficit          0.2677 ***        0.1573 ***
  (% of GDP)                    (0.000)           (0.000)
Total debt (% of GDP)                              0.0366 ***
                                                  (0.000)
Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability

Military

Law and order

Sargan test                     47.71             44.17
                                (0.89)            (0.19)
No. autocorrelation             -1.50             -1.54
                                (0.13)            (0.12)
No. of observations            432               432

                                   Model 5           Model 6

Inflation.L1                     0.3961 ***        0.3215 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.3591 ***       -0.3824 ***
                                (0.000)           (0.000)
Remittance growth                0.0004            0.0016
                                (0.853)           (0.448)
Log of nominal exchange rate    -4.3954 ***       -5.8327 ***
                                (0.000)           (0.000)
Trade (% of GDP)                 0.1287 ***        0.0887 ***
                                (0.000)           (0.000)
Current account deficit          0.1785 ***        0.1960 ***
  (% of GDP)                    (0.000)           (0.000)
Total debt (% of GDP)            0.0489 ***        0.0506 ***
                                (0.000)           (0.000)
Agricultural output              0.1676 ***
  (% of GDP)                    (0.001)
Crude oil prices growth (%)                       -0.0044
                                                  (0.105)
U.S. interest rate

Democracy

Government stability

Military

Law and order

Sargan test                     47.48             40.41
                                (0.66)            (0.43)
No. autocorrelation             -1.49             -1.61
                                (0.14)            (0.11)
No. of observations            432               432

                                   Model 7           Model 8

Inflation.L1                     0.3917 ***        0.3719 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.2959 ***       -0.3599 ***
                                (0.000)           (0.000)
Remittance growth                0.0013            0.0029
                                (0.472)           (0.124)
Log of nominal exchange rate    -4.1302 ***       -4.7221 ***
                                (0.000)           (0.000)
Trade (% of GDP)                 0.1039 ***        0.0872 ***
                                (0.000)           (0.000)
Current account deficit          0.1736 ***        0.1864 ***
  (% of GDP)                    (0.000)           (0.000)
Total debt (% of GDP)            0.0354 ***        0.0211 ***
                                (0.000)           (0.006)
Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate               0.1608 ***
                                (0.000)
Democracy                                         -1.1206 ***
                                                  (0.000)
Government stability

Military

Law and order

Sargan test                     45.86             44.11
                                (0.22)            (0.18)
No. autocorrelation             -1.58             -1.57
                                (0.12)            (0.12)
No. of observations            432               432

                                   Model 9          Model 10

Inflation.L1                     0.3004 ***        0.4028 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.5576 ***       -0.3290 ***
                                (0.000)           (0.000)
Remittance growth               -0.0002           -0.0008
                                (0.934)           (0.699)
Log of nominal exchange rate    -4.4777 ***       -3.8424 ***
                                (0.001)           (0.000)
Trade (% of GDP)                 0.0051            0.1157 ***
                                (0.757)           (0.000)
Current account deficit          0.1673 ***        0.1478 ***
  (% of GDP)                    (0.001)           (0.000)
Total debt (% of GDP)            0.0315 ***        0.0356 ***
                                (0.000)           (0.000)
Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability            -0.2944 **
                                (0.008)
Military                                          -0.6837 *
                                                  (0.018)
Law and order

Sargan test                     37.66             45.18
                                (0.75)            (0.23)
No. autocorrelation             -1.69             -1.53
                                (0.09)            (0.12)
No. of observations            432               432

                                  Model 11

Inflation.L1                     0.3830 ***
                                (0.000)
GDP growth (%)                  -0.3812 ***
                                (0.000)
Remittance growth               -0.0019
                                (0.360)
Log of nominal exchange rate    -4.8458 ***
                                (0.000)
Trade (% of GDP)                 0.1114 ***
                                (0.000)
Current account deficit          0.1493 ***
  (% of GDP)                    (0.000)
Total debt (% of GDP)            0.0401 ***
                                (0.000)
Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability

Military

Law and order                    0.4225
                                (0.277)
Sargan test                     45.71
                                (0.12)
No. autocorrelation             -1.61
                                (0.11)
No. of observations            432

p-values are in parentheses. *, **, and *** denote significance at
10%, 5%, and 1% levels of significance,  respectively. L1 stands for
one-period lag.

Table 5. System GMM Results (log[1 + (inf/100)] as Dependent Variable)

                                   Model 1           Model 2

Log inflation.L1                 0.5364 ***        0.5428 ***
                                (0.000)           (0.000)
GDP growth (%)                   0.0006            0.0005
                                (0.113)           (0.416)
Remittance growth                0.0001 ***        0.0001 ***
                                (0.000)           (0.000)
Log of nominal exchange         -0.0016 **        -0.0004
  rate                          (0.013)           (0.667)
Trade (% of GDP)                                   0.0003 ***
                                                  (0.000)
Current account deficit
  (% of GDP)
Total debt (% of GDP)

Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability

Military

Law and order

Sargan test                     51.25             41.58
                                (0.80)            (0.22)
No. autocorrelation             -0.93             -0.70
                                (0.35)            (0.49)
No. of observations            486               486

                                   Model 3           Model 4

Log inflation.L1                 0.5378 ***        0.5058 ***
                                (0.000)           (0.000)
GDP growth (%)                   0.0015 **         0.0004
                                (0.008)           (0.479)
Remittance growth                0.0001 ***        0.0001 *
                                (0.000)           (0.044)
Log of nominal exchange         -0.0031 ***       -0.0043 ***
  rate                          (0.000)           (0.000)
Trade (% of GDP)                 0.0002 ***       -0.0001
                                (0.000)           (0.106)
Current account deficit         -0.0001            0.0002
  (% of GDP)                    (0.737)           (0.169)
Total debt (% of GDP)                              0.0002 ***
                                                  (0.000)
Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability

Military

Law and order

Sargan test                     43.45             50.26
                                (0.65)            (0.55)
No. autocorrelation             -0.81             -1.02
                                (0.42)            (0.31)
No. of observations            486               486

                                   Model 5           Model 6

Log inflation.L1                 0.4727 ***        0.5105 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.0008 *          0.0006
                                (0.046)           (0.124)
Remittance growth                3.36e-06          0.0001 *
                                (0.721)           (0.032)
Log of nominal exchange         -0.0029 ***       -0.0054 ***
  rate                          (0.000)           (0.000)
Trade (% of GDP)                -0.0001           -0.0001 *
                                (0.400)           (0.098)
Current account deficit          0.0005 ***        0.0003 *
  (% of GDP)                    (0.001)           (0.058)
Total debt (% of GDP)            0.0002 ***        0.0001 ***
                                (0.000)           (0.000)
Agricultural output              0.0012 ***
  (% of GDP)                    (0.000)
Crude oil prices growth (%)                        0.0001
                                                  (0.323)
U.S. interest rate

Democracy

Government stability

Military

Law and order

Sargan test                     49.69             46.83
                                (0.42)            (0.28)
No. autocorrelation             -1.19             -1.08
                                (0.23)            (0.28)
No. of observations            486               486

                                   Model 7           Model 8

Log inflation.L1                 0.4906 ***        0.5132 ***
                                (0.000)           (0.000)
GDP growth (%)                   0.0001            0.0007
                                (0.904)           (0.151)
Remittance growth                0.0001 ***        0.0001 ***
                                (0.005)           (0.007)
Log of nominal exchange         -0.0053 ***       -0.0032 ***
  rate                          (0.000)           (0.000)
Trade (% of GDP)                -0.0001           -0.0002 **
                                (0.405)           (0.018)
Current account deficit          0.0005 ***       -0.0001
  (% of GDP)                    (0.002)           (0.704)
Total debt (% of GDP)            0.0002 ***        0.0001 ***
                                (0.000)           (0.000)
Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate               0.0011 *
                                (0.014)
Democracy                                          0.0019
                                                  (0.144)
Government stability

Military

Law and order

Sargan test                     46.55             43.82
                                (0.72)            (0.81)
No. autocorrelation             -1.07             -1.12
                                (0.29)            (0.26)
No. of observations            486               486

                                   Model 9          Model 10

Log inflation.L1                 0.5083 ***        0.5091 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.0006           -0.0003
                                (0.356)           (0.559)
Remittance growth                0.0001 *          3.94e-06
                                (0.041)           (0.763)
Log of nominal exchange         -0.0047 ***       -0.0044 ***
  rate                          (0.000)           (0.000)
Trade (% of GDP)                -0.0001           -0.0001 *
                                (0.273)           (0.099)
Current account deficit          0.0002            0.0002
  (% of GDP)                    (0.143)           (0.368)
Total debt (% of GDP)            0.0002 ***        0.0002 ***
                                (0.000)           (0.000)
Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability             0.0003
                                (0.655)
Military                                          -0.0038 ***
                                                  (0.001)
Law and order

Sargan test                     38.33             48.48
                                (0.39)            (0.71)
No. autocorrelation             -1.12             -1.19
                                (0.26)            (0.23)
No. of observations            486               486

                                  Model 11

Log inflation.L1                 0.5276 ***
                                (0.000)
GDP growth (%)                  -0.0009
                                (0.366)
Remittance growth                0.0001
                                (0.186)
Log of nominal exchange         -0.0066 ***
  rate                          (0.000)
Trade (% of GDP)                -0.0001
                                (0.442)
Current account deficit          0.0001
  (% of GDP)                    (0.927)
Total debt (% of GDP)            0.0002 ***
                                (0.000)
Agricultural output
  (% of GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability

Military

Law and order                   -0.0021
                                (0.296)
Sargan test                     47.33
                                (0.29)
No. autocorrelation             -0.96
                                (0.34)
No. of observations            486

p-values are in parentheses. *, **, and *** denote significance at
10%, 5%, and 1% levels of significance,  respectively. L1 stands for
one-period lag.

Table 6. Differenced GMM Results (log[1 + (inf/100)] as Dependent
Variable)

                                   Model 1           Model 2

Log inflation.L1                 0.3859 ***        0.4236 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.0050 ***       -0.0041 ***
                                (0.000)           (0.000)
Remittance growth                0.0001           -9.22e-06
                                (0.124)           (0.238)
Log of nominal exchange         -0.0389 ***       -0.0297 ***
  rate                          (0.000)           (0.000)
Trade (% of GDP)                                   0.0008 ***
                                                  (0.000)
Current account deficit (%
  of GDP)
Total debt (% of GDP)

Agricultural output (% of
  GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability

Military

Law and order

Sargan test                     38.35             47.51
                                (0.50)            (0.72)
No. autocorrelation             -1.21             -1.10
                                (0.23)            (0.27)
No. of observations            432               432

                                   Model 3           Model 4

Log inflation.L1                 0.4192 ***        0.4005 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.0032 ***       -0.0023 ***
                                (0.000)           (0.000)
Remittance growth               -2.18e-06          5.00e-6
                                (0.847)           (0.761)
Log of nominal exchange         -0.0517 ***       -0.0447 ***
  rate                          (0.000)           (0.000)
Trade (% of GDP)                 0.0009 ***        0.0008 ***
                                (0.000)           (0.000)
Current account deficit (%       0.0021 ***        0.0015 ***
  of GDP)                       (0.000)           (0.000
Total debt (% of GDP)                              0.0003 ***
                                                  (0.002)
Agricultural output (% of
  GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability

Military

Law and order

Sargan test                     51.87             47.69
                                (0.94)            (0.87)
No. autocorrelation             -1.24             -1.22
                                (0.22)            (0.22)
No. of observations            432               432

                                   Model 5           Model 6

Log inflation.L1                 0.3884 ***        0.3481 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.0031 ***       -0.0031 ***
                                (0.000)           (0.000)
Remittance growth               -0.0001            5.64e-06
                                (0.220)           (0.807)
Log of nominal exchange         -0.0357 ***       -0.0490 ***
  rate                          (0.000)           (0.000)
Trade (% of GDP)                 0.0011 ***        0.0008 ***
                                (0.000)           (0.000)
Current account deficit (%       0.0011 ***        0.0016 ***
  of GDP)                       (0.000)           (0.000)
Total debt (% of GDP)            0.0004 ***        0.0004 ***
                                (0.000)           (0.000)
Agricultural output (% of        0.0014 ***
  GDP)                          (0.009)
Crude oil prices growth (%)                       -0.0001 **
                                                  (0.010)
U.S. interest rate

Democracy

Government stability

Military

Law and order

Sargan test                     42.96             35.56
                                (0.75)            (0.34)
No. autocorrelation             -1.19             -1.43
                                (0.23)            (0.15)
No. of observations            432               432

                                   Model 7           Model 8

Log inflation.L1                 0.3827 ***        0.3964 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.0020 ***       -0.0021 ***
                                (0.000)           (0.000)
Remittance growth               -7.46e-06          5.71e-06
                                (0.695)           (0.743)
Log of nominal exchange         -0.0397 ***       -0.0504 ***
  rate                          (0.000)           (0.000)
Trade (% of GDP)                 0.0009 ***        0.0009 ***
                                (0.000)           (0.000)
Current account deficit (%       0.0015 ***        0.0019 ***
  of GDP)                       (0.000)           (0.000)
Total debt (% of GDP)            0.0003 ***        0.0002 ***
                                (0.004)           (0.000)
Agricultural output (% of
  GDP)
Crude oil prices growth (%)

U.S. interest rate               0.0011 **
                                (0.011)
Democracy                                         -0.0090 ***
                                                  (0.000)
Government stability

Military

Law and order

Sargan test                     41.05             48.90
                                (0.56)            (0.43)
No. autocorrelation             -1.32             -1.28
                                (0.19)            (0.20)
No. of observations            432               432

                                   Model 9          Model 10

Log inflation.L1                 0.3273 ***        0.4100 ***
                                (0.000)           (0.000)
GDP growth (%)                  -0.0040 ***       -0.0024 ***
                                (0.000)           (0.000)
Remittance growth               -0.0001           -0.0001
                                (0.062)           (0.078)
Log of nominal exchange         -0.056 ***        -0.0383 ***
  rate                          (0.000)           (0.000)
Trade (% of GDP)                -0.0001            0.0012 ***
                                (0.483)           (0.000)
Current account deficit (%       0.0018 ***        0.0013 ***
  of GDP)                       (0.000)           (0.000)
Total debt (% of GDP)            0.0003 **         0.0004 ***
                                (0.018)           (0.000)
Agricultural output (% of
  GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability            -0.0012
                                (0.070)
Military                                           0.0001
                                                  (0.980)
Law and order

Sargan test                     38.60             45.00
                                (0.25)            (0.77)
No. autocorrelation             -1.44             -1.28
                                (0.15)            (0.20)
No. of observations            432               432

                                  Model 11

Log inflation.L1                 0.3903 ***
                                (0.000)
GDP growth (%)                  -0.0026 ***
                                (0.000)
Remittance growth               -0.0001 **
                                (0.009)
Log of nominal exchange         -0.0502 ***
  rate                          (0.000)
Trade (% of GDP)                 0.0009 ***
                                (0.000)
Current account deficit (%       0.0015 ***
  of GDP)                       (0.000)
Total debt (% of GDP)            0.0004 ***
                                (0.000)
Agricultural output (% of
  GDP)
Crude oil prices growth (%)

U.S. interest rate

Democracy

Government stability

Military

Law and order                   -0.0017
                                (0.603)
Sargan test                     42.10
                                (0.33)
No. autocorrelation             -1.37
                                (0.17)
No. of observations            432

p-values are in parentheses. *, **, and *** denote significance at
10%, 5%, and 1% levels of significance,  respectively. L1 stands for
one-period lag.