Long-term growth in Europe: what difference does the crisis make?

Crafts, Nicholas

OECD projections for European countries imply that the crisis will have no long-term effect on trend growth. An historical perspective says this is too optimistic. Not only is the legacy of public debt and its requirement for fiscal consolidation unfavourable but the experience of the 1930s suggests that much needed supply-side reforms are now less probable--indeed policy may well become less growth friendly. Whereas the 1940s saw the Bretton Woods agreement and the Marshall Plan pave the way for the 'Golden Age', it is unlikely that anything similar will rescue Europe this time around.

Keywords: Financial crisis; fiscal consolidation; growth projections; Marshall Plan; productivity growth; supply-side reform

JEL Classifications: N14; 052

Introduction

About the turn of the century it became conventional wisdom that economic growth in western European countries was a bit disappointing and that a rapid decline in Europe's share of world GDP was underway. Nevertheless, there seemed no reason to doubt the plausibility of a scenario of 'business as usual' underpinned by gradual structural reform delivering growth at around 2 per cent per year for the foreseeable future.

Obviously, the financial crisis which erupted in 2008 was a rude shock to expectations and many people thought it signalled significant policy reforms. On the maxim of 'never waste a good crisis', optimists could see this as an opportunity to overcome obstacles to reform which were holding back trend growth. In contrast, pessimists could think that a combination of misdiagnoses and political pressures might push policymakers in the direction of interventions which would undermine long-run growth. Economic history offers examples of both--the former might appeal to the birth of the European Golden Age in the dark days of the late 1940s while the latter could point to the implications of the policy responses to the Great Depression of the 1930s.

Econometric analyses of the implications of financial crises for growth offer some pointers but leave open how far European growth prospects may have been or may yet be adversely affected by the financial crisis. It is already generally agreed that a substantial permanent fall in the level of output is likely, although there is some uncertainty about its magnitude. It is less clear whether trend growth will be reduced through declines in the rate of productivity growth. Clearly, one way in which this might happen is that major crises can lead to significant changes in supply-side policy. Here it is useful to turn to economic history to complement what can be learnt from econometrics.

In the light of this challenge, this paper considers European growth performance and its likely trajectory in the absence of the crisis. Then, the direct effects of the crisis on growth are assessed. The evidence points to a significant negative effect in the absence of policy reforms that improve productivity growth. Are these policy reforms likely to be made? Historical experience, notably in the context of the 1930s, suggests that the crisis will produce strong pressures in the opposite direction. Overall, although this is not yet recognised by agencies like the OECD, it seems likely that the crisis implies a significantly reduced average European growth rate over the period to 2030.

Growth before the crisis

It is well-known that Western European growth performance was lack-lustre from the mid-1990s up to the start of the current financial crisis. This was a far cry from the so-called Golden Age of the early postwar years and for most countries was a period of falling behind rather than catching up with the United States even though it appeared that, on the whole, supply-side policy had improved. Yet, while American productivity growth accelerated, European productivity growth slowed down.

The period from the early 1950s to the mid-1990s was an era when Western Europe was catching up with the United States (table 1). During the Golden Age (195073), real GDP per person (Y/P) and labour productivity (measured either per worker (E) or per hour worked(HW)) both grew much faster in most European countries than in the United States. In the following period of growth slowdown (1973-95), labour productivity continued to grow faster than in the United States but catch-up in real GDP per person ceased. The discrepancy is, of course, explained by slower growth in labour inputs in European countries as unemployment rose and work-years shortened; this is also captured in table 1.

From 1995 to the eve of the current crisis, real GDP per person in Western Europe declined slowly relative to the United States; rather than catching up, Europe had started to fall behind. Table 1 shows that for the EU15 the ratio was 67.5 per cent in 2007 compared with 70.0 per cent in 1995. The data in this table also show that the main reason was slower labour productivity growth in Europe. Trends in annual hours worked were now more similar while the earlier tendency for employment rates to fall relative to the United States was reversed, so that total hours worked per person rose in Europe.

The growth rate of real GDP per hour worked increased in the United States between 1973-95 and 1995-2007 from 1.28 per cent per year to 2.05 per cent per year. In contrast, in the EU15 it fell from 2.69 per cent per year to 1.17 per cent per year. The rate of labour productivity growth fell in most European countries. Indeed, labour productivity growth in Italy and Spain was below 1 per cent per year after 1995. By contrast, Sweden saw a productivity revival, while for part of the period Ireland continued to be a Celtic Tiger and both countries exceeded the American productivity growth rate. So while there was falling behind in productivity performance on average, there was also considerable diversity in European performance.

The acceleration in American productivity growth was underpinned by ICT. Historical comparisons reveal that the impact of ICT has been relatively large and also that it has come through very quickly. This new general purpose technology (GPT) is unprecedented in its rate of technological progress, reflected in the speed and magnitude of the price falls in ICT equipment (Crafts, 2004). The main impact of ICT on economic growth comes through its diffusion as a new form of capital equipment rather than through total factor productivity (TFP) growth in the production of ICT equipment. This is because users get the benefit of technological progress through lower prices and as prices fall more of this type of capital is installed. (1)

The implication is that ICT has offered Europe a great opportunity to increase its productivity growth. However, the estimates of the contribution of ICT capital deepening to the growth of labour productivity in table 2 show that European countries have been less successful than the United States in seizing this opportunity. That said, ICT production has boosted productivity growth, notably in Finland, Ireland and Sweden, and the use of ICT capital has made a strong contribution, especially in the services sector, in countries like the UK. Table 2 suggests that strong productivity performance in the recent past relied on one or both of ICT production and market services.

The empirical evidence is that the diffusion of ICT has been aided by complementary investments in intangible capital and high-quality human capital but weakened by relatively strong regulation in terms of employment protection and restrictions to competition, especially in the distribution sector (Conway et al., 2006). Since these forms of regulation have weakened over time, the story is not that European regulation has become more stringent but rather that existing regulation became more costly in the context of a new technological era. Of course, European countries have varied considerably in these respects; for example, the UK and Sweden have been better placed than Italy and Spain.

The example of ICT prompts some more general comments on European supply-side policies in the decades before the crisis. In some respects, these improved in terms of providing conditions favourable to growth. For example, European countries generally became more open to trade, with positive effects on productivity partly as a result of the European single market. Over time, years of schooling have steadily increased and product market regulation that inhibits competition has been reduced. Corporate tax rates have fallen since the early 1980s. The UK, in particular, benefited from the big increase in competition in product markets from deregulation and abandoning protectionism that took place from the 1970s through the 1990s (Crafts, 2012a). Nevertheless, supply-side policies were in need of reform prior to the crisis if the issue of disappointing growth performance was to be adequately addressed. This was the case quite generally but especially in Southern Europe.

Thus, there have been serious question marks about the quality of schooling in many European countries which recent research suggests exacts a growth penalty. A measure of cognitive skills, based on test scores, correlates strongly with growth performance (Hanushek and Woessmann, 2012) and it is striking that even the top European countries, such as Finland, have fallen behind Japan and South Korea with some countries, such as Germany and, especially, Italy, deteriorating. These authors' estimates are that, if cognitive skills in Greece were at the standard of South Korea, its long-run growth would be raised by about 1 percentage point per year. Woessmann et al. (2007) show that the variance in outcomes in terms of cognitive skills is explained by the way the schooling system is organised (for example, with regard to the role of external examinations and how much competition between providers is allowed) rather than educational spending.

Competition and competition policy has tended to be weaker than in the United States. This has raised markups and lowered competitive pressure on managers to invest and to innovate with adverse effects on TFP growth (Buccirossi et al., 2013; Griffith et al., 2010). Clearly, productivity growth in market services was very disappointing in many European countries (table 2). One reason is continued weakness of competition reflected in high price-cost mark-ups which have survived the introduction of the Single Market (Hoj et al., 2007). Addressing these issues by reducing the barriers to entry maintained by member states would have raised productivity performance significantly but governments still have considerable discretion to maintain these barriers notwithstanding the Services Directive (Badinger and Maydell, 2009). It should also be noted that failure to deal with excessive regulation in professional services in particular has had adverse effects on productivity growth in user industries (Barone and Cingano, 2011).

Finally, research into the impact of fiscal policy on growth has suggested that the structure of taxation has significant effects. Generally speaking, direct taxes are more damaging than indirect taxes. The substantial increase in social transfers in European countries in the later decades of the 20th century was financed to a considerable extent by 'distortionary' taxation; the estimates of Kneller et al. (1999) indicate that the average 10 percentage point increase in the share of direct tax revenues in GDP between 1965 and 1995 entailed a fall in the growth rate of about 1 percentage point. Financing this expansion of government outlays by a different tax mix would have been considerably better for growth.

On the eve of the crisis, there was widespread agreement on reforms which would improve Western European growth performance, although, of course, the extent of what was needed varied across countries. This consensus was based on empirical analysis of the experience of recent decades and reflected the discussion of this section. The very influential analysis by Sapir (2006) stressed the importance at the EU level of completing the Single Market in services and at the national level of reforming labour market and social policies where these reduced flexibility and employment.

OECD economists, who believe that implementation of these reforms has been made more urgent by the crisis, provide some useful quantification of the possible benefits from structural policy reforms which is summarised in table 3. Drawing on a series of OECD studies of the effects of policies on income levels and growth rates, Barnes et al. (2011) sum up by proposing improvements to the quantity and quality of education, strengthening competition, cutting unemployment benefits, reducing and reforming taxes, and lowering employment protection. These would either raise the growth rate or in some cases provide a transitional boost to growth as the economy moves to higher employment and output levels. The authors claim that addressing all policy weaknesses by moving up to the OECD average level has a potential GDP gain of 10 per cent for the average country after ten years and 25 per cent eventually. (2)

Growth implications of the financial crisis

This section considers the direct implications of the crisis rather than those which might come via a wider array of policy responses that affect the progress of supply-side reforms; these will be addressed in the following section.

An obvious starting point is to consider what economists who make long-term projections have to say. The team at Goldman Sachs have reviewed their BRICs projections which offer comparisons of future developed market and emerging market growth (Wilson et al., 2011). Their view is essentially that the crisis will have no long-term impact and they project real GDP growth in Europe at 2.2 per cent per year for each of the decades 2010-19 and 2020-29--identical to performance in the 1990s.

A more detailed analysis is provided by OECD (2012, ch. 4). Some of the main projections are summarised in table 4. The OECD team explicitly state that their basic assumption is that the crisis will have a small effect on the level of potential real GDP in the OECD area of about 2.5 per cent but no effect on the trend rate of growth. Currently, output levels in OECD countries are below potential but these 'output gaps' are assumed to disappear over the next few years. The analysis assumes that the crisis does not affect the pace of structural reforms of the kind listed above which will make a modest contribution on average to future growth. In particular, it is noticeable that the labour productivity growth projections in table 4 for the Euro Area and for the aggregate of OECD countries are quite optimistic and call for a stronger performance between 2012 and 2030 than for 1995-2007. Even troubled Eurozone economies will generally share in this experience. Slower employment growth in future will, however, hold real GDP growth below the pre-crisis rate.

It is well known that financial crises can have permanent adverse direct effects on the level and possibly also the trend growth rate of potential output. Indeed, this is a major reason why such crises have serious fiscal implications, including big increases in structural deficits as a percentage of GDP (Reinhart and Rogoff, 2009). Thinking in terms of a production function or growth accounting, there may be direct adverse effects on capital inputs (K) as investment is interrupted, on human capital (HK) if skills are lost or restructuring makes them redundant, on labour inputs (HW) through increases in equilibrium unemployment, and on TFP if R&D is cut back or innovative firms cannot get finance. Given that in the long-run TFP growth is the fundamental underpinning of the rate of trend growth, the key question is whether this is affected by financial crises.

Modern econometric studies are not quite conclusive. Furceri and Mourougane (2009) estimate that for OECD countries a severe banking crisis reduces the level of potential output by about 4 per cent while the review of the evidence in IMF (2009, ch. 4), which also covers lower-income economies, suggests 10 per cent, with the level of capital, labour inputs and TFP each accounting for about a third of this. In both papers long-run trend growth is found not to be affected but in both cases the confidence intervals are large and the transition period while the levels effect materialises may be quite long. If the level of potential output suffers an adverse shock, then for a country like the UK the structural fiscal deficit is likely to rise by a similar amount as a share of GDP (IFS, 2010) and fiscal consolidation may be required to head off unstable debt dynamics.

A celebrated historical example of a severe banking crisis is the United States during the Great Depression of the 1930s when about one in three banks failed and financial intermediation was severely disrupted with severe consequences for investment and GDP (Bernanke, 1983). What does that experience reveal? The obvious feature of the 1930s, reflected in table 5, is that the financial crisis undermined growth in the capital stock. Had growth of the capital to labour ratio continued at the pre-1929 rate, by 1941 it would have been about 25 per cent larger and, accordingly, potential GDP per hour worked perhaps 8 per cent bigger. Growth of labour inputs was sluggish, impaired perhaps by the impact of the New Deal. However, TFP growth was very strong, powered by sustained R&D, and Field (2013) has labelled the 1930s the most technologically progressive decade of the twentieth century in the United States building on the strong fundamentals in place from before the Great Depression. Overall, the clear impression is that the result of the banking crisis was to lower the level of productive potential rather than its growth rate, an outcome in line with the OECD's assumptions. (3)

For today's European countries the analogue may be that continued TFP growth in ICT, which will further reduce the price of ICT equipment, has the scope to underpin future growth. Oulton (2012) estimates that even without reforms which would speed up its diffusion this will deliver 0.52 percentage points to future growth on average across the OECD; with reform, this could go up to 0.73 percentage points. Table 6 reports projections of the future scope for ICT to deliver productivity growth for European countries both with and without reform. The key point to note is that if the use of ICT capital were as important as in Sweden, the most successful adopter of ICT, then the ICT capital contribution would rise appreciably, notably by over 0.2 percentage points per year in the large continental economies. (4)

Banking crises can also be expected to affect future growth through their implications for regulation of the financial sector and through their fiscal legacy. With regard to regulation, there is general agreement that a major requirement is to reduce the leverage of the banking system to lower the chances of future crises. This will generally entail the banks having more equity capital to absorb losses which will raise the cost of capital to the banks and to their borrowers. However, the evidence suggests that, although this will reduce the capital stock and the level of GDP in the future, the effect will probably be small. Miles et al. (2013) provide an illustrative calculation which suggests that halving leverage from 30 to 15 might cost a little under 0.2 per cent of GDP. It should also be recognised that growth did not seem to be impaired during most of the 20th century when leverage was much lower.

An obvious implication of the financial crisis is that it will leave a legacy of much increased ratios of public debt to GDP (D/Y). Partly this will come through the costs of recapitalising banks but mainly it will come through the borrowing by governments as a result of the financial crisis induced recession and the fall in tax receipts that this has entailed. The median increase in D/Y in advanced countries following a banking crisis in the recent past is estimated to have been 21 per cent of GDP (Laeven and Valencia, 2012). (5) OECD countries are more vulnerable on this score than in the 1930s because they entered the crisis with higher D/Y ratios (Ali Abbas et al., 2011) and during the crisis have provided some fiscal stimulus and have not tried to over-ride the automatic stabilisers.

The long-term implications of substantial increases in D/Y are likely to be unfavourable for growth, as is highlighted by growth models of the overlapping-generations variety. The adverse impacts can occur through a number of transmission mechanisms including reductions in market-sector capital formation, higher long-term interest rates and higher tax rates. Empirical research on advanced economies has found negative relationships; for example, Kumar and Woo (2010) estimate that a 10 percentage point increase in D/Y is associated with a fall of about 0.2 percentage points in growth. It has also been suggested that the adverse effect on growth is non-linear and rises sharply once the debt ratio reaches a critical level which has recently been claimed to be around 90 per cent of GDP (Checherita and Rother, 2010). OECD (2012) notes that, in the Euro Area, United States and OECD as a whole, D/Y is approaching 100 per cent and points out that if the results of these papers are taken literally then this could reduce future trend growth by between 0.5 and 0.75 percentage points. (6)

High levels of D/Y create further worries about future fiscal sustainability which add to the pressures for increases in taxes and/or cuts in public expenditure, i.e., fiscal consolidation. Recent work by OECD economists in the wake of the crisis has given some guidance on the size of the adjustments that may be needed. For example, if it is desired to achieve D/Y = 60 per cent by 2030 (the Maastricht Treaty rule), then the required average improvement in the primary balance from 2011-30 would have been 7.0 per cent of GDP in the UK, about 7.6 per cent in Ireland and the United States, and 8.6 and 9.7 per cent respectively in Portugal and Greece (OECD, 2012). (7)

Continuing fiscal consolidation is unlikely to be expansionary; on the contrary, the implications are likely to be deflationary and to entail (possibly considerable) GDP losses. The estimates in Guajardo et al. (2011) are that, on average, a fiscal consolidation of 1 per cent of GDP reduces real GDP by 0.62 per cent over the following two years. This can be mitigated if an offsetting monetary-policy stimulus is possible and/or the exchange rate depreciates. (8) If the fiscal adjustment is achieved through expenditure cuts rather than tax increases and accompanied by structural reforms, the evidence is that output losses may be lower, in particular because private investment tends to respond favourably (Alesina et al., 2012). Nevertheless, it is reasonable to suppose that post-crisis fiscal adjustment is likely to be a drag on medium-term growth, especially for countries inside the euro. (9)

It is generally believed that expenditure-based consolidations have a greater chance of success (Molnar, 2012) and it might be thought that if this argument informs post-crisis policy it would minimise harmful supply-side effects on growth by mitigating against distortionary tax increases. (10) However, cuts in expenditure on education (which adds to human capital) and on infrastructure (which adds to physical capital) are bad for long-term growth. Unfortunately, previous episodes of fiscal stringency have been notable for their negative impact on public investment (Mehrotra and Valila, 2006). Moreover, it is noticeable that, at high levels of debt, addressing a rising D/Y typically entails cuts in both public investment and education spending (Bacchiocchi et al., 2011). The strong likelihood that post-crisis fiscal consolidation will undermine these expenditures does not bode well for the growth prospects of highly-indebted EU countries. (11)

Policy responses to the crisis: lessons from the 1930s

If medium-term trend growth rates are significantly reduced as a result of the crisis, a major reason could well be the broad economic policy response. Given the magnitude of the problems that Europe now confronts, it is useful to consider the lessons from the 1930s, a period notable for policy changes which helped short term economic and political objectives but damaged long-term growth performance.

First, it is well-known that the Great Depression saw big increases in protectionism, partly reflected in the tariff rates reported in table 7. Intra-European trade costs rose by an average of about 20 per cent between 1929 and 1938 (cf. table 10). Protectionism accounted for about 40 per cent of the fall in trade volumes during the Great Depression (Irwin, 2012). The most interesting analysis of this protectionism is by Eichengreen and Irwin (2010) who show that, on average, countries which devalued had lower tariffs. They argue that protection in the 1930s is best seen as a second-best policy which was used when the conventional macroeconomic tools, fiscal and monetary policy, were unavailable. The countries to which this description most obviously applies today are Euro Area economies with sovereign debt and competitiveness problems; a likely implication is that they will be even less inclined to move towards implementing the Single Market in services.

However, obstacles to the use of monetary and fiscal policy are more pervasive than this. On the monetary side, with interest rates at the zero lower bound (ZLB) conventional monetary policy is circumscribed. (12) On the fiscal side, worries about fiscal sustainability have already undermined willingness to use fiscal stimulus. This can be expected to continue to be the case in future, perhaps reinforced by new fiscal rules for countries in the Euro Area. Moreover, the legacy of the crisis will be a lengthy period when public debt to GDP ratios are at a level which potentially renders fiscal stimulus ineffective or possibly even counterproductive. (13) Auerbach and Gorodnichenko (2011) find that at debt to GDP ratios greater than 100 per cent fiscal multipliers are close to zero even in deep recessions, while Ilzetzki et al. (2010) suggest that, on average, the fiscal multiplier is zero on impact and in the long run is negative at debt to GDP ratios above 60 per cent.

This makes protectionist policies more likely and we have already seen a lurch in this direction, albeit not to the extent of the rampant protectionism of the 1930s. Table 8 reports policy interventions recorded by Global Trade Alert. These are mostly not flagrant violations of WTO rules and traditional tariff measures are only a small part of what has happened. Thus, 84 per cent of interventions in the EU have employed policy instruments that are subject to low or no regulation by the WTO using measures such as bailouts and subsidies with the EU state-aids regime effectively suspended (Aggarwal and Evenett, 2012). So, although research has suggested that tariff barriers contributed little (perhaps only 2 per cent) to the downturn (Kee et al., 2013), it would be misleading to conclude that the protectionist threat is negligible. Indeed, a WTO-compatible protectionist wave is a real possibility (Evenett and Vines, 2012).

Second, the 1930s saw a general retreat from competition in the advanced countries together with increases in regulation and nationalisation. Voters were less willing to trust in markets and wanted greater state intervention. This was reflected in policy developments at the time and in the postwar settlements of the 1940s. Both the British and American governments sought to encourage cartelization--the difference being that in the United States the Supreme Court struck down the provisions of the National Industrial Recovery Act in 1935. In the United Kingdom, competition in product markets was greatly weakened and this lasted long into the postwar period. In the late 1950s, tariffs were still at mid-1930s levels and about 60 per cent of manufacturing output was cartelised. The retreat from competition had adverse effects on productivity performance over several decades and provided the context in which industrial relations problems and sleepy management proliferated (Crafts, 2012a). In the United States, product market regulation was substantially increased with adverse implications for economic efficiency and it was not until the 1970s and 1980s that this was reformed (Vietor, 1994).

In Europe, the response was also to embrace selective industrial policy, picking winners and supporting national champions but especially helping losers, a reaction which was intensified when macroeconomic troubles and globalisation challenges returned in the 1970s (Foreman-Peck, 2006). Across Europe, state ownership was extended so that countries typically entered the Golden Age with nationalised industries supplying 10 per cent of GDP (Millward, 2011). In practice, industrial policy was heavily skewed to slowing down exit with adverse consequences for productivity performance and this may be an inherent characteristic of such policies (Baldwin and Robert-Nicoud, 2007). By the 1980s, selective industrial policy deservedly had a bad name (Geroski and Jacquemin, 1985) and allowing creative destruction to take its course can be seen to have been a superior policy for high-income countries (Fogel et al., 2008). Unfortunately, in the aftermath of the present crisis, the early signs are not good; there has already been a serious reversion to the use of selective industrial policy and this has come in the guise of helping losers (Aggarwal and Evenett, 2012).

Third, given the pressures on public finances which the crisis has triggered, as in the 1930s and 1940s, there may be a serious move to financial repression with a view to reducing the interest costs of debt service and holding down the real interest rate relative to the growth rate, thereby lowering the required primary surplus for fiscal sustainability. The political attractions of an alternative to fiscal contraction are apparent. Such a policy involves limiting in various ways the free flow of capital between countries, i.e., reducing the integration of capital markets.

Ali-Abbas et al. (2011) found that big debt reductions in the period 1945-70 were, unusually, characterised by a much larger component from the growth-interest differential rather than from primary surpluses, while Wyplosz (2001) showed that capital controls and credit restraints were instrumental in significantly lowering European real interest rates at this time. (14) There was, however, a cost to European growth from reducing the efficiency of capital markets which Voth (2003) estimated as at least 0.7 percentage points per year. The increment to growth from greater financial integration in European countries which was realised pre-crisis (Gehringer, 2013) is at risk if moves towards financial repression intensify and entail barriers to capital flows together with reductions in cross-border lending. (15)

Fourth, the 1930s' experience suggests that a strategy of devaluation and sovereign default could be an attractive escape route from the euro for the periphery countries of Southern Europe because it allows more policy sovereignty and a route to return to growth. The key point is that devaluation and default were good for growth. Indeed, there is a very clear correlation between the countries that exited early from the Gold Standard and those that recovered rapidly from the slump (Bernanke, 1995), see table 9. Abandoning the Gold Standard allowed countries to reduce both nominal and real interest rates, to relax fiscal policy, and to escape the need to reduce money wages and prices through a prolonged recession in order to regain international competitiveness. It is also clear that sovereign default, which improved the fiscal arithmetic by eradicating debt overhangs and bolstered the balance of payments through the elimination of debt-service flows, promoted short term growth (Eichengreen and Portes, 1990).

The decision to leave the Gold Standard was analysed by Wolf (2008) who used an econometric model to examine the odds of remaining in this fixed exchange rate system. The model accurately predicts departures and shows that a country was more likely to exit from the Gold Standard if its main trading partner had done so, if it had returned to gold at a high parity, if it was a democracy, or if the central bank was independent. Conversely, it was less likely to leave if it had large gold reserves, less price deflation, and strong banks. In other words, the loss of international competitiveness and greater pain from deflationary pressures hastened a country's exit.

Thus, the 1930s' experience suggests that once one country exits, others may quickly follow. The pressures on the survival of the euro are likely to intensify. How costly this might be is a matter of speculation but estimates have been made; for example, Cliffe (2011) suggests that after five years real GDP in the Euro Area would still be 5 per cent lower than at the break-up. It should be recognised, however, that the benefit/cost ratio of leaving the Gold Standard then was very different from that of leaving the euro today; a decision to reintroduce a national currency now might engender 'the mother of all financial crises' (Eichengreen and Temin, 2010) through instantaneous capital flight and a collapse of the banking system.

It is also reasonable to suggest that the demise of the currency union would have a permanent adverse effect on GDP levels, although perhaps not as large as has sometimes been claimed. The currency-union effect on trade volumes was once thought to be very large but better econometrics and the opportunity to examine the actual impact of EMU now suggests that trade volumes increased by perhaps 2 per cent (Baldwin et al., 2008) with the implication that the trade effect on GDP was less than 1 per cent. There are, however, several channels through which EMU may have raised productivity and a recent study found that EMU had raised the level of real GDP per hour worked by 2 per cent (Barrell et al., 2008); this would potentially be at risk.

Fifth, the shock of the 1930s encouraged workers to demand greater social protection and promoted tighter regulation of the labour market. In the United States this was famously addressed by the New Deal. There, the 1930s saw the federal government pass the Social Security Act in 1935, which established a wide range of benefits including unemployment insurance and retirement benefits. Another long-lasting intervention was the Fair Labor Standards Act of 1938 which brought in minimum wages and overtime restrictions (Fishback, 2007). Across European countries we see the development of much more ambitious social policies which leave their footprint through a big increase of social transfers between 1930 and 1950 (Crafts, 2012b). This t exemplifies the well-known result established by Rodrik (1998) that in the subsequent return to globalisation t voters demanded greater government spending to cushion exposure to external shocks. To the extent that this was financed by distortionary taxation, there was a negative implication for growth.

The overall message of this section is that there is a serious threat both to levels of potential output and to t medium-term trend growth from policy responses to the pressures created by the crisis. Moreover, the general direction of the 1930s' policy response to economic crisis runs very much in the opposite direction to the supply-side reforms which are required to speed up European growth. At best, this does not bode well for the agenda of completing the single market and making labour markets more flexible and employment-friendly put forward by Sapir (2006). At worst, there may be adverse policy moves which undermine medium-term growth performance.

If this gloomy prospectus is correct, then European growth performance will probably undershoot the 'business-as-usual' projections in OECD (2012). Nevertheless, OECD (2013) is optimistic about the future of structural reforms, as it stresses that the crisis has seen a significant increase in the responsiveness of governments to OECD proposals for actions to improve supply-side policies. In the short term, to a large extent, this is a corollary of strong pressures to restore competitiveness and improve fiscal sustainability. This is, however, probably not a good guide to the long term, especially if there is a 'lost decade'.

In the early 1930s, the initial response of the British government to its fiscal problems included cuts in unemployment benefits and public sector wages but this was merely a blip on the road to 1945 when voters expressed their strong disapproval of the market economy. Across Europe in the 1930s, prolonged stagnation significantly increased the electoral prospects of right-wing extremist parties (de Bromhead et al., 2012) which were not market-friendly. In this context, not only might it be reasonable to worry about recent election results but it should also be recognised that opinion polls show disappointingly low support for the market economy in countries where economic recovery seems a remote prospect. (16) This suggests that optimism about supply-side policy based on early responses to the crisis should not be overdone.

Reconstructing Europe: lessons from the 1940s

At the end of World War II, many thought that the prospects for European growth were quite gloomy, yet The 'Golden Age' was just around the corner. This was phase of rapid catch-up growth during which Western Europe greatly reduced the massive productivity gap (cf. table 1) with the United States that had emerged in the previous 40 years (Crafts and Toniolo, 2008). Clearly, this episode cannot be repeated but it is useful to consider changes in the international policy regime, in particular the Bretton Woods agreement and the Marshall Plan, which were conducive to the start of the Golden Age and European economic integration, with regard to the implications for today.

The Bretton Woods agreement meant that the typical OECD country moved to a macroeconomic trilemma choice of fixed exchange rates, independent monetary policy and capital controls (Obstfeld and Taylor, 2004). This was designed, with the GATT, to promote a return to freer international trade after the trade wars of the 1930s. Table 10 reflects some success in this regard. The prevalence of capital controls can also be seen as a choice within the 'political trilemma' posited by Rodrik (2000), reproduced in figure 1, which states that it is generally only possible to have at most two of the following: deep economic integration, democratic politics and the nation state.

The 1930s' implosion of the Gold Standard can be understood in terms of this political trilemma, as follows. In the 1920s, with the return to the Gold Standard, countries had signed up to the 'golden straitjacket', which had been acceptable in the context of very limited democracy in the 19th century but in the 1930s democratic politics at the level of the nation state overruled this policy choice. The point is that to retain the benefits of deep economic integration would have required action to organise it through democratic politics at a supranational level. (17) So, when reconstruction of the international economy was subsequently undertaken under the auspices of the 1944 Bretton Woods agreement, economic integration was severely restricted by controls on international capital flows (the 'Bretton-Woods compromise' in figure 1).

[FIGURE 1 OMITTED]

It might be suggested that the initial design of the euro sought to impose an equivalent to the 'golden straightjacket'. History suggests that it will be difficult to continue this strategy under the pressures brought on by the crisis and in today's world it is neither feasible nor desirable to seek to reinstate the capital controls of the 1950s. If so, then logic points to a different solution to the political trilemma problem based either on a modern equivalent to the 1950s' 'Bretton-Woods Compromise' or the 'global federalism' of a 'United States of Europe'.

The idea of the 'Bretton-Woods Compromise' was to sacrifice some aspects of economic integration to provide sufficient policy space to make saving the remaining aspects politically acceptable. Obviously, this came at a cost--European growth in the Golden Age would have been even stronger without capital controls. It is arguable that we are already seeing an equivalent in terms of the return of selective industrial policy, creeping protectionism and moves towards financial repression. If so, this will also incur a cost in lower growth over the medium term.

The alternative, 'United States of Europe', solution would require major institutional reform within the EU and would entail banking union, fiscal union, and a constitution that ended Europe's 'democratic deficit'. Wolf (2012) spells out what this might enable. He notes that it would allow more effective European fiscal and monetary policy at the ZLB, and political legitimisation of a much higher level of transfer payments from an expanded European budget while also finding a way to share burdens of adjustment between surplus and deficit countries. However, delivering this will be difficult and, at best, it may take years to achieve the success that would preserve the income gains that have accrued from European integration and preclude creeping protectionism in capital and product markets.

The Marshall Plan was a major programme of aid which transferred $12.5 billion from the United States to Western Europe during the years 1948 to 1951 and provided inflows to recipient countries which typically averaged about 2 per cent of GDP per year. (18) It helped reduce the costs of the Bretton-Woods Compromise by speeding up European integration via trade liberalisation and the conditionality that it entailed also promoted pro-market reforms that were conducive to growth. As De Long and Eichengreen (1993) stressed, rather than being a handout, the Marshall Plan was a "structural adjustment program" along the lines of the Washington Consensus--"the most successful ever"--which succeeded by raising productivity growth and steered Europe away from becoming Argentina. It worked by tipping the balance in favour of pro-market structural reforms that raised productivity growth rather than through a direct stimulus. (19)

In particular, each country signed a bilateral treaty with the United States which committed them to follow policies of financial stability and trade liberalisation while the Organization for European Economic Cooperation (OEEC) provided 'conditional aid' to back an intra West European multilateral payments agreement; in 1950, recipients had to become members of the European Payments Union (EPU). This lasted until 1958, by which time intra-European trade was 2.3 times that of 1950 and a gravity-model analysis confirms that the EPU had a large positive effect on trade levels. (20) This can be seen as a stepping stone to further trade liberalisation through increases in the political clout of exporting firms relative to import-competing firms (Baldwin, 2006). The long-term effect of economic integration raised European income levels substantially, by nearly 20 per cent by the mid-1970s according to estimates by Badinger (2005). (21)

This suggests that there might be something to be gained from a 'Marshall Plan' as a way of mitigating the costs of the modern equivalent of a Bretton-Woods Compromise and as an insurance policy to reduce the likelihood of the disintegration of the euro in the interval before the federalist solution is feasible but also as a means to improve growth prospects in southern Europe. The objective of a 'Real' Marshall Plan for Southern Europe would be to underpin European economic integration and the survival of the euro by incentivising supply-side reforms to increase productivity growth. (22) The central component, as in the 1940s, would be to formulate a successful structural adjustment programme with strict conditionality focused on improving productive potential rather than simply spending more of the EU budget. The programme would include product market reforms, including serious moves to implement fully the Single Market, fiscal reforms to broaden the tax base and switch towards consumption and property taxes, labour-market reforms to increase flexibility and to reduce the equilibrium level of unemployment. Such changes would not only improve productivity (cf. table 3) but also go some way towards restoring lost competitiveness. These are economic-policy changes rather than investments in projects and, generally, incur political rather than monetary costs. In the longer term, improvements in educational quality should also be a focal point.

Unfortunately, it seems most unlikely that a new Marshall Plan of this kind is feasible. To be credible, the funds would have to be committed but only released when reforms had been implemented satisfactorily (Allard and Everaert, 2010)--similar to the deal that worked in the context of EU enlargement in 2004. Whether the EU could implement this has to be doubtful. Moreover, the general record of structural adjustment programmes administered by the IMF and World Bank is that the results have been disappointing both in terms of compliance and outcomes. More specifically, the success or failure of such programmes depended heavily on domestic political economy considerations and willingness to embrace reform, i.e., finding 'good candidates' to support (Dollar and Svensson, 2000). It would surely be difficult at present to persuade Northern Europeans that Southern European countries are 'good candidates' or that non-compliance would be effectively punished and the policy prescriptions would not be welcomed by Southern European voters. Regrettably, therefore, it seems that trying this antidote to problems in the Euro Area is most unlikely although, in principle, it could play a very useful role in underpinning growth as did the Marshall Plan of the 1940s.

Conclusions

The legacy of the crisis is a number of significant downside risks to long-term European growth which do not yet seem to be widely understood and have not yet been incorporated into projections made by agencies like the OECD.

A major implication of the crisis is that average public debt to GDP ratios in European countries will approach or even exceed 100 per cent. Past experience says that the existence of such high debt ratios will quite possibly reduce growth performance by 0.5 percentage points or more per year. Attempts to address this issue through fiscal consolidation which may last for many years will further depress growth. Fiscal consolidation can be undertaken in a variety of ways with very different implications for growth potential; past experience suggests that there is a real danger that government expenditure on education and infrastructure will be jeopardised.

Supply-side reforms undertaken now could potentially raise European growth rates perhaps by as much as 0.5 to 1 percentage point per year over the period to 2030. The general thrust of these reforms would include fiscal changes to reduce distortionary taxation, strengthening competition and reducing regulations that damage productivity, and improving the quality of education. Generally speaking, this could be done without undermining public finances but not without upsetting many voters. The debt overhang makes reforms that raise the growth rate in European countries more urgent because of their favourable implications for fiscal sustainability but, unfortunately, less likely. Indeed, the pressures of the crisis may well push policy in adverse directions, especially if there seems to be no scope to stimulate economic activity using conventional fiscal or monetary policies. The 1930s' crisis led to more regulation, less competition, increased protectionism and financial repression.

The experience of the 1930s also suggests that there is a real risk of European economic integration being partially reversed or even of the euro collapsing in chaos with a large cumulative GDP loss over five or more years. Devaluation and sovereign default may have increasing political appeal. The logic of the political trilemma points to a fully-federal Europe as the way to avert these outcomes but this is unlikely to be achieved easily or quickly. Unlike the 1940s, there seems no real prospect of a Marshall Plan to help rescue the situation.

Conventional wisdom is still that medium-term growth prospects are unaffected by the crisis which began in 2007. This seems too optimistic. Considering both the direct effects and the pressures to change policies in directions which will undermine rather than stimulate growth, it is very possible that real GDP growth in the Euro Area between 2012 and 2030 will be well below the OECD's projections.

REFERENCES

Aggarwal, V.K. and Evenett, S.J. (2012), 'Industrial policy choice during the crisis era', Oxford Review of Economic Policy, 28, pp. 261-83.

Alesina, A., Favero, C. and Giavazzi, F. (2012), 'The output effects of fiscal consolidation', NBER Working Paper No. 18336.

Ali Abbas, S.M., Belhocine, N., El-Ganainy, A. and Horton, M. (2011), 'Historical patterns and dynamics of public debt- evidence from a new database', IMF Economic Review, 59, pp. 717-42.

Allard, C. and Everaert, L. (2010), 'Lifting euro area growth: priorities for structural reforms and governance', IMF Staff Position Note 10/19.

Auerbach, A. and Gorodnichenko, Y. (2011), 'Fiscal policy in recession and expansion', NBER Working Paper No. 17447.

Bacchiocchi, E., Borghi, E. and Missale, A. (2011), 'Public investment under fiscal constraints', Fiscal Studies, 32, pp. 11-42.

Badinger, H. (2005), 'Growth effects of economic integration: evidence from the EU member states', Review of World Economics, 14 I, pp. 50-78.

Badinger, H. and Maydell, N. (2009), 'Legal and economic issues in completing the EU internal market for services: an interdisciplinary perspective', Journal of Common Market Studies, 47, pp. 693-717.

Baldwin, R. (2006), 'Multilateralising regionalism: spaghetti bowls as building blocks on the path to global free trade', The World Economy, 29, pp. 1451-518.

Baldwin, R., Di Nino, V., Fontagne, L., De Santis, R. and Taglioni, D. (2008), 'Study on the impact of the Euro on trade and foreign direct investment', European Economy Economic Papers No. 321.

Baldwin. R.E. and Evenett, S.J. (2012), 'Beggar-thy-neighbour policies during the crisis era: causes, constraints and lessons for maintaining open borders', Oxford Review of Economic Policy, 28, pp. 211-34.

Baldwin, R.E. and Robert-Nicoud, F. (2007), 'Entry and asymmetric lobbying: why governments pick losers', Journal of the European Economic Association, 5, pp. 1064-93.

Barnes, S., Bouis, R., Briard, P., Dougherty, S. and Eris, M. (2011), 'The GDP impact of reform: a simple simulation framework', OECD Economics Department Working Paper No. 834.

Barone, G. and Cingano, F. (2011), 'Service regulation and growth: evidence from OECD countries', Economic Journal, 121, pp. 931-57.

Barrell, R., Gotschalk S., Holland, D., Khoman, E., Liadze, I. and Pomerantz, O. (2008), 'The impact of EMU on growth and employment', European Economy Economic Papers No. 318.

Ben-David, D., Lumsdaine, R. and Papell, D. H. (2003), 'Unit roots, postwar slowdowns and long-run growth: evidence from two structural breaks', Empirical Economics, 28, pp. 303-19.

Bernanke, B. (1983), 'Non-monetary effects of the financial crisis in the propagation of the great depression', American Economic Review, 73, pp. 257-276.

--(1995), 'The macroeconomics of the great depression: a comparative approach', Journal of Money, Credit and Banking, 27, pp. 1-28.

Bernanke, B. and James, H. (1991) 'The gold standard, deflation, and financial crisis in the great depression: an international comparison', in Hubbard, R.G. (ed.), Financial Markets and Financial Crises, Chicago, University of Chicago Press, pp. 33-68.

Blanchard, O. and Leigh, D. (2013), 'Growth forecast errors and fiscal multipliers', IMF Working Paper No. 13/1.

Buccirossi, P., Ciari, L., Duso, T., Spagnolo, G. and Vitale, C. (2013), 'Competition policy and productivity growth: an empirical assessment', Review of Economics and Statistics, forthcoming.

Checherita, C. and Rother, P. (2010), 'The impact of high and growing government debt on growth: an empirical investigation for the Euro area', ECB Working Paper No. 1237.

Cliffe, M. (2011), EMU Break-Up: Pay Now, Pay Later, London, ING. Conway, P., de Rosa, D., Nicoletti, G. and Steiner, F. (2006), 'Regulation, competition and productivity convergence', OECD Economics Department Working Paper No. 509.

Crafts, N. (2004), 'Steam as a general purpose technology: a growth accounting perspective', Economic Journal, 114, pp. 338-51.

--(2011), 'The Marshall Plan: a reality check', University of Warwick CAGE Working Paper No. 49.

--(2012a), 'British relative economic decline revisited: the role of competition', Explorations in Economic History, 49, pp. 17-29.

--(2012b), 'Western European growth prospects: an historical perspective', University of Warwick CAGE Working Paper No. 72.

--(2012c), 'Saving the Eurozone: is a 'real' Marshall Plan the answer?', CAGE-Chatham House Policy Paper No. 1.

Crafts, N. and Toniolo, G. (2008), 'European economic growth, 1950-2005: an overview', CEPR Discussion Paper No. 6863.

De Bromhead, A., Eichengreen, B. and O'Rourke, K. H. (2012), 'Right-wing political extremism in the great depression', NBER Working Paper No. 17871.

De Long, J.B. and Eichengreen, B. (1993), 'The Marshall Plan: history's most successful adjustment program', in Dornbusch, R., Nolling, W. and Layard, R. (eds), Postwar Economic Reconstruction and Lessons for the East Today, Cambridge, Mass., MIT Press, pp. 189-230.

Dollar, D. and Svensson, J. (2000), 'What explains the success or failure of structural adjustment programmes?', Economic Journal, 110, pp. 894-917.

Eichengreen, B. (1993), Reconstructing Europe's Trade and Payments, Manchester, Manchester University Press.

Eichengreen, B. and Irwin, D. (2010), 'The slide to protectionism in the great depression: who succumbed and why?', Journal of Economic History, 70, pp. 871-97.

Eichengreen, B. and Portes, R. (1990), 'The interwar debt crisis and its aftermath', World Bank Research Observer, 5, pp. 69-94.

Eichengreen, B. and Temin, P. (2010), 'Fetters of gold and paper', Oxford Review of Economic Policy, 26, pp. 370-84.

Eichengreen, B. and Uzan, M. (1992), 'The Marshall Plan: economic effects and implications for Eastern Europe and the former USSR', Economic Policy, 14, pp. 13-75.

Evenett, S.J. and Vines, D. (2012), 'Crisis-era protectionism and the multilateral governance of trade: an assessment', Oxford Review of Economic Policy, 28, pp. 195-210.

Fernandez-Villaverde, J, Guerron-Quintana, P. and Rubio-Ramirez, J.F. (2011), 'Supply-side policies and the zero lower bound', NBER Working Paper No. 17543.

Field, A.J. (2013), 'Economic growth and recovery in the United States, 1919-41', in Crafts, N. and Fearon, P. (eds), The Great Depression of the 1930s: Lessons for Today, Oxford, Oxford University Press, pp. 358-94.

Fishback, P.V. (2007), 'The New Deal', in Fishback, P.V. (ed.), Government and the Economy: a New History, Chicago, University of Chicago Press, pp. 384-430.

Fogel, K., Morck, R. and Yeung, B. (2008), 'Big business stability and economic growth: is what's good for General Motors good for America?', Journal of Financial Economics, 89, pp. 83-108.

Foreman-Peck, J. (2006), 'Industrial policy in Europe in the 20th century', EIB Papers, II (I), pp. 37-62.

Furceri, D. and Mourougane, A. (2009), 'The effect of financial crises on potential output: new empirical evidence from OECD Countries', OECD Economics Department Working Paper No. 699.

Gehringer, A. (2013), 'Growth, productivity and capital accumulation: the effects of financial liberalization in the case of European integration', International Review of Economics and Finance, 25, pp. 291-309.

Geroski, P. and Jacquemin, A. (1985), 'Industrial change, barriers to mobility and European industrial policy', Economic Policy, I, pp. 170-218.

Grifflth, R., Harrison, R. and Simpson, H. (2010), 'Product market reform and innovation in the EU', Scandinavian Journal of Economics, 112, pp. 389-415.

Guajardo, J., Leigh, D. and Pescatori, A. (2011), 'Expansionary austerity: new international evidence', IMF Working Paper No. 11/158.

Hagemann, R.P. (2012), 'Fiscal consolidation part 6: what are the best policy instruments for fiscal consolidation?', OECD Economics Department Working Paper No. 937.

Hanushek, E.A. and Woessmann, L. (2012), 'Do better schools lead to more growth? Cognitive skills, economic outcomes, and education', Journal of Economic Growth, 17, pp. 267-321.

Hoj, J., Jimenez, M., Maher, M., Nicoletti, G. and Wise, M. (2007), 'Product market competition in OECD countries: taking stock and moving forward', OECD Economics Department Working Paper No. 575.

Holland, D. (2012), 'Less austerity, more growth?', NIESR Discussion Paper No. 400.

Ilzetzki, E., Mendoza, E.G. and Vegh, C.A. (2010), 'How big (small?) are fiscal multipliers?', NBER Working Paper No. 16479.

Institute for Fiscal Studies (2010), Green Budget 2010.

International Monetary Fund (2009), World Economic Outlook.

Irwin, D.A. (2010), 'Did France cause the great depression?', NBER Working Paper No. 16350.

--(2012), Trade Policy Disaster, Cambridge, Mass., MIT Press.

Jacks, D.S., Meissner, C.M. and Novy, D. (2011), 'Trade booms, trade busts, and trade costs', Journal of International Economics, 83, pp. 185-201.

Kamps, C. (2005), 'Is there a lack of public capital in the European Union?', EIB Papers, 10(I), pp. 73-93.

Kee, H.L., Neagu, C. and Nicita, A. (2010), 'Is protectionism on the rise?: assessing national trade policies during the crisis of 2008', Review of Economics and Statistics, forthcoming.

Kneller, R., Bleaney, M. and Gemmell, N. (1999), 'Fiscal policy and growth: evidence from OECD countries', Journal of Public Economics, 74, pp. 171-90.

Kumar, M.S. and Woo, J. (2010), 'Public debt and growth', IMF Working Paper No. 10/174.

Laeven, L. and Valencia, F. (2012), 'Systemic banking crises database update', IMF Working Paper No. 12/163.

Larch, M. and Turroni, A. (2011), 'Received wisdom and beyond: lessons from fiscal consolidation in the EU', National Institute Economic Review, 217, pp. R1-R18.

Maddison, A. (2003), The World Economy: Historical Statistics, Paris, OECD.

Mehrotra, A. and Valila, T. (2006), 'Public investment in Europe: evolution and determinants in perspective', Fiscal Studies, 27, pp. 443-71.

Miles, D., Yang, J. and Marcheggiano, G. (2013), 'Optimal bank capital', Economic Journal, 123, pp. 1-37.

Millward, R. (2011), 'Public enterprise in the modern western world: an historical analysis', Annals of Public and Cooperative Economics, 82, pp. 375-98.

Molnar, M. (2012), 'Fiscal consolidation part 5: what factors determine the success of consolidation efforts?', OECD Economics Department Working Paper No. 936.

Obstfeld, M. and Taylor, A. (2004), Global Capital Markets: Integration, Crisis and Growth, Cambridge, Cambridge University Press.

OECD (2012), Economic Outlook.

--(2013), Economic Policy Reforms 2013: Going for Growth. Oulton, N. (2012), 'Long-term implications of the ICT revolution: applying the lessons of growth theory and growth accounting', Economic Modeling, 29, pp. 1722-36.

Pew Research (2012), 'Pew Global Attitudes Project' at http://www.pewglobal.org.

Prados de la Escosura, L., Roses, J. and Sanz Villaroya, 1. (2011), 'Economic reforms and growth in Franco's Spain', Revista de Historia Economica, 30, pp. 45-89.

Reinhart, C.M. (2012), 'The return of financial repression', CEPR Discussion Paper No. 8947.

Reinhart, C.M. and Rogoff, K.S. (2009), This Time is Different: Eight Centuries of Financial Folly, Princeton, Princeton University Press.

Reinhart, C.M. and Sbrancia, M.B. (2011), 'The liquidation of government debt', NBER Working Paper No. 16893.

Rodrik, D. (1998), 'Why do open economies have bigger governments?', Journal of Political Economy, 106, pp. 997-1032.

--(2000), 'How far will international economic integration go?", Journal of Economic Perspectives, 14(I), pp. 177-86.

Sapir, A. (2006), 'Globalization and the reform of European social models', Journal of Common Market Studies, 44, pp. 369-90.

Timmer, M., Inklaar, R., O'Mahony, M. and van Ark, B. (2010), Economic Growth in Europe: a Comparative Economic Industry Perspective, Cambridge, Cambridge University Press.

Vietor, R.H.K. (1994), 'Contrived competition: economic regulation and de-regulation, 1920-1980s', Business History, 36(4), pp. 1-32.

Voth, H.-J. (2003), 'Convertibility, currency controls and the cost of capital in Europe, 1950-1999', International Journal of Finance and Economics, 8, pp. 255-76.

Wilson, D., Trivedi, K., Carlson, S. and Ursua, J. (2011), 'The BRICs I 0 years on: halfway through the great transformation', Goldman Sachs Global Economics Paper No. 208.

Woessmann, L., Ludemann, E, Schutz, M. and West, M.R. (2007), 'School accountability, autonomy, choice and the level of student achievement: international evidence from PISA 2003', OECD Education Working Paper No. 13.

Wolf, N. (2008), 'Scylla and Charybdis: explaining Europe's exit from gold, January 1928 to December 1936', Explorations in Economic History, 45, pp. 383-401.

--(2012), 'Crises and policy responses within the political trilemma: Europe, 1929-36 and 2008-2011', EHES Working Papers in Economic History No. 16.

Wyplosz, C. (200 I), 'Financial restraints and liberalization in postwar Europe', in Caprio, Honohan, G.P. and Stiglitz, J. (eds), Financial Liberalization: How Far, How Fast?, Cambridge, Cambridge University Press, pp. 125-58.

Notes

(1) In a country with no ICT production, a neoclassical growth model whose Cobb-Douglas production function has two types of capital (ICT and other) shows that the steady state rate of growth will be TFP growth plus a term denoting the rate of real price decline for ICT capital multiplied by the share of ICT capital in national income, all divided by labour's share of national income (Oulton, 2012).

(2) Some reforms, notably to educational systems, take a long time to pay off.

(3) Time-series econometrics can also be used to address this issue (although the picture is muddied by World War II) and this also points to a levels decrease--actually followed by a modest increase in trend growth (Ben-David et al., 2003).

(4) Table 6 shows that for most countries it is ICT use rather than ICT production that dominates the contribution that this GPT makes to growth.

(5) In some cases such as Ireland (73 per cent), the increase has been much greater than this.

(6) The OECD projections in table 4 do not incorporate this prediction.

(7) The basic algebra related to stabilising the debt ratio is that Ad = b + (i - [pi] - [DELTA]Y/Y)d where d is the debt/GDP ratio, b is the primary budget deficit, i.e., the budget deficit without including interest payments on the debt, i is the nominal interest rate, ]pi] is the rate of inflation and [DELTA]Y/Y is the rate of growth of real GDP. So for [increment of d] = 0, the required primary budget surplus -b = d(i - [pi] - [DELTA]Y/Y). Price deflation means that [pi] is negative and this clearly makes the fiscal task much harder given that i cannot be negative. On the other hand if [DELTA]Y/Y > (i - [pi]), i.e., if growth is greater than the real interest rate it is possible to stabilise (or even reduce) d while running a primary deficit. Normally, the primary balance has to take the strain but, interestingly, this was not the case during the European Golden Age when in the era of rapid catch-up and financial repression the growth rate was much greater than the real interest rate (All Abbas et al., 2011). We return to the implications of this point in the next section.

(8) This strategy is, of course, not available for Euro Area countries,

(9) It should be noted that such effects are not taken into account in the OECD projections reported in table 4.

(10) The validity of this claim for the recent experience of EU countries is disputed by Larch and Turroni (2011).

(11) Pre-crisis, many EU countries were already investing too little to maintain the stock of public capital at a growth-maximising level (Kamps, 2005).

(12) If, however, the policy response at the ZLB is for reforms that raise productivity and wealth, their impact on recovery is enhanced because it is not offset by interest rate increases (Fernandez-Villaverde et al., 2011).

(13) None of this should be taken to imply that fiscal multipliers were low at the start of the crisis--on the contrary they were probably greater than I in many countries at the ZLB and with a credit crunch, see Blanchard and Leigh (2013) and Holland (2012).

(14) For the UK, it has been estimated that the financial repression 'tax' yielded 3.6 per cent of GDP per year between 1945 and 1980 (Reinhart and Sbrancia, 2011).

(15) There have already been some in the form of financial regulation (Reinhart, 2012) while cross-border investment in Europe has fallen by over 50 per cent since 2008.

(16) In response to the question 'Are people better off in a free market economy?' in 2012 only 44 per cent in Greece, 47 per cent in Spain and 50 per cent in Italy agreed (Pew Research, 2012). In 2007, 67 per cent in Spain and 73 per cent in Italy had agreed (no data for Greece).

(17) It should also be noted that enduring prolonged recessions, perhaps in the attempt to comply with the golden straitjacket, helped spawn the rise of extreme right-wing political parties although the structure of the electoral system and the depth of democratic traditions in determining political outcomes also mattered (de Bromhead et al., 2012).

(18) More details of how it worked can be found it Crafts (2011).

(19) Eichengreen and Uzan (1992) estimate that the direct effects on growth averaged only 0.3 per cent per year.

(20) See Eichengreen (1993) which also contains a detailed description of the logic and mechanics of EPU.

(21) Exclusion from the Marshall Plan and EPU postponed but did not necessarily preclude trade liberalisation, as is shown by the 1959 reform in Spain which raised the growth rate by 1 percentage point per year for the next decade and a half (Prados de la Escosura et al., 2011).

(22) This is surely not what those who have called for a new Marshall Plan have in mind; their focus is typically on income transfers not conditionality-based reform. For a fuller discussion of the idea of a 'Real Marshall Plan', see Crafts (2012c).

Nicholas Crafts, Warwick University. E-mail: N.Crafts@warwick.ac.uk. An earlier version of this paper was presented at the inaugural CEPR-Modena conference, 'Growth in Mature Economies', in November 2012. I am grateful to participants for very helpful comments. The paper has also benefited from suggestions made by Dawn Holland, Simon Kirby and two anonymous referees. Remaining errors and omissions are my responsibility.

Table 1. Output and productivity in Europe and the
United States, 1950-2007

a) Growth rates (% per year)

 Real GDP
 Real GDP Real GDP /hour
 Real GDP /person /worker worked

1950-73

EU 15 5.97 4.05 4.15 4.80
USA 3.92 2.45 2.30 2.56

1973-95

EU 15 2.25 1.89 2.00 2.69
USA 2.86 1.80 1.12 1.28

1995-2007

EU 15 2.58 1.80 0.92 1.17
USA 3.17 2.11 1.92 2.05

b) Decomposition of EU 15/USA real GDP/person gap, 1950-2007

 Y/P Y/HW HW/E E/P

1950 0.482 0.381 1.190 1.063
1973 0.680 0.629 1.092 1.000
1995 0.700 0.853 0.974 0.843
2007 0.675 0.769 0.947 0.928

Source: Derived from The Conference Board Total Economy
Database (2012).
Note: The table shows the identity Y/P = Y/HW x HW/E x E/P for
the ratio of EU 15/USA.

Table 2. Labour productivity growth in the market sector,
1995-2005 (% per year)

a) Growth accounting

 Non- Labour
 Labour ICTK/ ICT TFP productivity
 quality HW K/HW growth

Ireland 0.2 0.4 2.1 1.8 4.5
Sweden 0.3 0.6 I.I 1.6 3.6
UK 0.5 0.9 0.4 0.8 2.6
France 0.4 0.4 0.4 0.9 2.1
Portugal 0.2 0.6 1.3 -0.3 1.8
Germany 0.1 0.5 0.6 0.4 1.6
Spain 0.4 0.3 0.5 -0.8 0.4
Italy 0.2 0.3 0.5 -0.7 0.3
USA 0.3 1.0 0.3 1.3 2.9

b) Sectoral contributions

 Labour
 ICT Manu- Other Market productivity
 production facturing goods services growth

Ireland 1.0 2.2 0.2 1.4 4.5
Sweden I.I I.0 0.2 1.4 3.6
U K 0.5 0.5 0.2 1.6 2.6
France 0.4 0.7 0.3 0.7 2.1
Portugal 0.5 0.5 0.2 0.6 1.8
Germany 0.4 0.6 0.3 0.2 1.5
Spain 0.1 0.1 0.0 0.2 0.4
Italy 0.3 0.0 0.2 -0.1 0.3
USA 0.8 0.6 -0.1 1.8 2.9

Source: Timmer et al. (2010).
Note: Reallocation effects not reported

Table 3. Potential impact on real GDP per person of
structural policy reforms

 Product
 Labour market Edu- Total
 market Taxation regulation cation

Moving to OECD average

USA 0.3 1.4 0.0 2.5 4.2
France 4.5 10.9 2.2 2.1 19.7
Germany 6.1 9.9 0.0 0.0 16.0
UK 1.1 0.0 0.0 4.6 5.7
Sweden 6.5 6.4 0.0 0.1 13.0
Greece 6.0 10.1 22.0 5.8 43.9
Ireland 6.8 0.9 9.7 0.0 17.4
Italy 0.3 10.8 0.3 5.4 16.8
Portugal 7.3 0.7 8.5 21.8 38.3
Spain 3.5 4.6 0.0 6.3 14.4
10 per cent
reforms'
OECD
average 6.1 3.3 3.8 11.6 24.8

Source: Barnes et al. (2011).

Table 4. OECD long-term growth projections

 1995- 2008- 2012-17 2012-17
 2007 12 (actual) (potential) 2018-30

a) Real GDP growth (% per year)

OECD 2.8 0.5 2.4 2.1 2.3
Euro Area 2.3 -0.1 1.8 1.5 1.8
USA 3.1 0.6 2.7 2.1 2.4
France 2.2 0.1 2.2 1.8 2.1
Germany 1.6 0.6 1.7 1.6 1.1
U K 2.9 -0.5 1.9 1.5 2.1
Sweden 3.2 0.9 2.6 2.5 2.3
Greece 3.8 -2.8 1.7 0.6 2.4
Ireland 2.4 -1.7 2.6 1.2 2.6
Italy 1.5 -1.0 0.9 0.6 1.6
Portugal 3.2 -1.2 1.0 0.7 1.8
Spain 3.7 -0.4 2.3 1.5 2.3

b) Growth of Real GDP/worker (% per year)

OECD 1.7 0.5 2.1 2.3
Euro Area 1.0 0.3 1.5 1.8
USA 1.8 1.2 1.3 1.4
France 1.1 0.1 1.4 1.9
Germany 1.2 0.0 1.4 1.7
U K 1.9 -0.4 0.8 1.5
Sweden 2.4 0.4 1.8 1.9
Greece 2.5 -0.8 0.3 2.2
Ireland 2.3 1.2 0.9 1.3
Italy 0.3 -0.5 0.1 1.5
Portugal 1.4 0.2 0.5 1.7
Spain 0.1 1.7 0.8 1.6

Sources: 1995-2012: The Conference Board Total Economy
database; 2012-30: OECD (2012, ch. 4).

Table 5. Contributions to labour productivity growth in
the United States (% per year)

 K/HW HK/HW TFP Y/HW
 growth growth growth growth

1906-19 0.51 0.26 1.12 1.89
1919-29 0.31 -0.06 2.02 2.27
1929-41 -0.19 0.14 2.97 2.92
1941-48 0.24 0.22 2.08 2.54
1948-73 0.76 0.11 1.88 2.75

Source: Derived from Field (2013).
Note: Estimates are for private non-farm economy.

Table 6. ICT and long-run growth potential (% per year)

 ICT- ICT-use
 use own Swedish ICT
 [beta] [beta] output

Austria 0.46 0.76 0.22
Belgium 0.64 0.73 0.13
Czech Republic 0.53 0.81 0.27
Denmark 0.62 0.70 0.20
Finland 0.67 0.76 0.57
France 0.48 0.68 0.17
Germany 0.44 0.68 0.33
Hungary 0.58 0.79 0.44
Ireland 0.39 0.94 0.51
Italy 0.36 0.70 0.19
Netherlands 0.51 0.71 0.10
Slovenia 0.28 0.62 0.28
Spain 0.53 0.76 0.10
Sweden 0.70 0.70 0.24
UK 0.60 0.66 0.16
Unweighted average 0.52 0.73 0.26

Source: Oulton (2012).
Note: [beta] is the factor share of ICT
capital; a high value indicates relatively successful diffusion and
is conducive to a higher growth contribution. These projections are
based on a neoclassical growth model with 2 types of capital, ICT
capital and other capital, and a production function y =
[Ak.sub.NICT][alpha][k.sub.ICT][beta] where y is output per worker
and k denotes capital per worker. In the traditional model with one
type of capital, steady state labour productivity growth is
([DELTA]A/A)/ SL, where SL is labour's share of national income. In
the modified model, this is augmented by an additional term
(([beta][DELTA]p/p)/[s.sub.L] where [DELTA]p/p is the rate of
decline of the price of ICT capital goods relative to other capital
goods. [DELTA]p/p will reflect technological progress in the production
of ICT capital. The estimates assume that the real price of ICT
equipment continues to fall at 7 per cent per year.

Table 7. Tariff rates, 1928, 1935 and 1938 (%)

 1928 1935 1938

Austria 8.1 17.5 14.8
Belgium 3.4 8.3 6.7
Canada 15.9 15.4 13.9
Czechoslovakia 7.8 10.0 7.2
Denmark 5.5 8.2 7.3
France 6.6 16.9 16.6
Germany 7.9 30.1 33.4
Hungary 11.0 7.2 12.0
Italy 6.7 22.2 12.1
Japan 7.1 6.2 6.6
Netherlands 2.1 9.1 6.7
New Zealand 17.1 17.5 16.4
Norway 11.5 14.4 12.2
Spain 24.1 27.9 n/a
Sweden 9.3 10.1 9.5
Switzerland 9.3 23.3 18.1
United Kingdom 10.0 24.5 24.1
United States 13.8 17.5 15.5

Source: Eichengreen and Irwin (2010).
Note: Tariff rate is defined as customs revenue/value of imports.

Table 8. Crisis era protectionist measures recorded by
Global Trade Alert

Bailout/state aid 324 Migration 47
Trade defence 289 Investment 46
Tariff 166 Public procurement 41
Non-tariff barrier (n.e.s) 110 Export subsidy/incentives 38
Export taxes 85 Import ban 28

Source: Baldwin and Everett (2012).
Note: 'Trade defence' comprises antidumping, countervailing
duties and safeguard measures

Table 9. Dates of changes in gold standard policies and
economic recovery

 Return to 1929
 income level Devaluation

Austria 1939 09/1931
Belgium 1939 03/1935
Denmark (a) 09/1931
Finland 1934 10/1931
France 1939 10/1936
Germany 1935 (b)
Greece 1933 04/1932
Italy 1935 10/1936
Netherlands 1949 09/1936
Norway 1932 09/1931
Spain 1955 (c)
Sweden 1933 09/1931
Switzerland 1938 09/1936
United Kingdom 1934 09/1931
United States 1940 04/1933

Sources: Bernanke and James (1991); Maddison (2003).
Notes: (a) Real GDP per person never fell below the 1929 level in
Denmark. (b) Germany did not devalue but by imposing exchange
controls effectively left the gold standard in July 1931.
(c) Spain was not on the gold standard.

Table 10. Trade costs: Italy with various partner countries

 Germany UK Spain USA

1929 1.10 1.22 1.63 1.26
1950 1.27 1.36 2.40 1.40
1960 1.01 1.25 1.54 1.29
1970 0.79 1.21 1.42 1.22
1980 0.61 0.86 1.08 1.13
1990 0.56 0.84 0.87 1.13
2000 0.66 0.90 0.87 1.14

Source: Data underlying jacks et al. (2011) generously provided
by Dennis Novy.

Notes: Trade costs are inferred from a gravity model of trade
flows and comprise both policy and non-policy barriers to
trade.