What went wrong in Greece and how to fix it: Lessons for Europe from the Greek crisis
This column argues that the inevitable adjustment was more painful than necessary. The fiscal consolidation was too tight and too front-loaded, and, importantly, structural reform wasn’t properly sequenced. By concentrating on labour market rather than product market reforms, the sharp wage fall could not be paralleled by a similar reduction in prices, and now soaring inequality is undermining support for needed reforms.
Figure 1 shows the Greek real GDP since 1990, along with Germany’s and Italy’s, all normalised to 100 at the beginning of the period. The Greek GDP took off in the second half of the 90s. At its peak in 2008, it had grown by 65%, cumulatively. The following hard landing was equally impressive, but much more rapid. By 2013, GDP had fallen back to its 2000 level.
As it happens with asset price bubbles, the boom of the economy was not justified by underlying fundamentals, and therefore was unsustainable. Table 1 looks at the supply side and disentangles GDP growth into its main components. The share accounted by employment, employment composition, capital – split into Information and Communication Technology (ICT) and non-ICT capital – and Total Factor Productivity (TFP). During the ‘euro-boom’ years, the growth rate averaged 4.13% per year, and it was mostly explained by (non-ICT) investment. The equally dramatic fall since 2008, -4.45% per year on average, was largely due to labour shedding. What is important to note here is that from 1990 to 2007, TFP, the ‘engine’ of growth, was either falling or flat.1
Table 1. Average yearly contributions to GDP growth, percentages
Source: Author’s calculations from The Conference Board Total Economy Database (2015).
As a result, relative to Germany, between 1990 and 2008 Greece accumulated a 23% ‘productivity gap’ that persisted and increased in the bust period (see Figure 2). The consequence was a huge ‘competitiveness gap’ for the Greek economy. Figure 3 shows two measures of the real effective exchange rate, one based on unit labour costs (ULC) and one based on consumer prices (PCI). Between 1990 and 2009 the Greek economy experienced a 35% loss in competitiveness in terms of ULCs (slightly more than 20% in terms of prices), meaning that wages (and prices) rose much more rapidly than in its trading partners in the run up to the crisis.
Figure 2. Total factor productivity, normalised to 100 in 1990
Source: The Conference Board Total Economy Database (2015).
Figure 3. Real effective exchange rates, normalised to 100 in 2000Q1
Source: Zoltan Dervas, Bruegel , 2014.
To a large extent, the lack of productivity growth accounts for the hard landing of Greece. Figure 4 below considers the relationship between pre-crisis TFP growth and post-crisis GDP growth in 160 countries. Each country is represented by a dot. On the vertical axis I measure the country’s cumulative GDP growth since the beginning of the crisis (2008-13), while on the horizontal axis I measure the pre-crisis (2000-07) cumulative growth in TFP. Countries which experienced a high productivity growth before the crisis on average did better in terms of GDP growth afterwards.
Figure 4. Cumulative GDP growth 2008-2013 and TFP growth 2000-07
Source: Author’s calculations from The Conference Board Total Economy Database (2015).
Greece lies in the bottom left corner of the graph. Its low, pre-crisis growth of productivity makes it a good candidate for a large recession during the crisis. Yet, Greece lies well below the regression line (it is an outlier), implying that demand factors substantially contributed to the recession.
As productivity growth was flat, both the take-off and the hard-landing of the Greek economy were demand driven. Table 2 shows the familiar data for public debt, deficit, current account, consumption, investment, imports and exports for the pre-euro, euro-boom and euro bust periods, as a percentage of GDP. During the euro-boom period, there was a spree in consumption, investment and imports that was financed by public and foreign debt, while the real exchange rate appreciated; during the euro-burst, demand experienced a classical sudden stop along with a real depreciation.
Table 2. Expenditure and fiscal data
The hard landing was driven by the inability to access the capital markets, which forced a huge squeeze in private and public spending. Figure 5 below, borrowed from IMF (2013), shows the improvement in the fiscal primary balance that took place between 2009 and 2012. This was unprecedented compared to previous experiences in Europe, both in terms of size – about 9 percentage points of GDP – and intensity (three years). Since 2009, government revenues rose from 37 to 45% of GDP while public expenditures fell from above 50 to 45% (Figure 6). Particularly severe were cuts in public employment (Figure 7).
Figure 5. Fiscal adjustment
Sources: World Economic Outlook and IMF staff calculations.
Figure 6. Government revenue and expenditures as percentage of GDP
Source: Ameco (2015).
Figure 7. Public employment
Source: MAREG, Census database and Special Secretariat for State Owned Enterprises.
Far from obeying the ‘expansionary fiscal contraction’ paradigm,2 the fiscal consolidation in the context of widespread credit constraints and monetary union made the adjustment harsher than needed. Furthermore, the adjustment was not significantly helped by an ‘expenditure switch’ towards tradable goods. Exports rose, and imports fell (see Figure 8) but far from what would have been necessary to provide some cushion against the squeeze in domestic demand. Figure 9, borrowed from EC (2014), makes the point that exports of goods and, in particular, services, rose by much less compared to other European countries which went through more successful stabilisations (Ireland, Spain, Latvia, Portugal).
Figure 8. Exports and imports as percentage of GDP
Source: Ameco (2015).
Figure 9. Export-led recoveries
Source: EC (2014).
Competitiveness, wages and prices
In order to understand why, we must take another look at competitiveness in Figure 3. Since 2010, the real effective exchange rate measured by relative unit labour costs shows massive gains in competitiveness. In a couple of years the accumulated competitiveness gap was almost cancelled out, as the REER depreciated by about 35% since 1990. Price competitiveness, however – described by the CPI-based measure – improved by only a handful of percentage points. And exports depend on relative prices, not on relative wages. This sharp misalignment between real exchange rates in Greece can be partly explained by the fact that prices were falling in Greece’s trade partners as well as in Greece. But as Greek labour productivity was flat, the main reason was that nominal wages collapsed while prices were stable. Figure 10 shows the average nominal compensation per employee and the index of consumer prices, both normalised to 100 in 1990. Nominal wages boomed since entry in the euro, far exceeding price increases (and productivity increases). Following the bust in 2009 however, nominal wages crashed, but prices did not. In a few years, most of the gains in real wages obtained since 1990 were gone.
Figure 10. Prices and wages, normalised to 100 in 1990
Source: OECD (2015).
Clearly, the fact that exports did not rise ‘enough’ has other explanations beyond the modest gains in price competitiveness. Let me just mention a few:
- Most Greek firms are very small (well below 10 employees) and their size make them unable to access foreign markets;
- Small firms suffered an unprecedented credit crunch;
- Bureaucracy is an obstacle for obtaining export licenses;
- Exports are concentrated in low and medium-tech goods, such as Fuels, Metals, Food Products and Chemicals (see Figure 11), and econometric evidence points to their low price-inelasticity (see Athanaoglu et al 2010); and, Many among Greece’s main trading partners (Turkey, Italy, Germany, Cyprus, Bulgaria, and the UK, in descending order) underwent sever recessions.
Figure 11. Export composition
Source. World Bank (2015).
In addition to the implications for competitiveness, the sharp decline in real wages had two other important consequences – it further depressed demand in the presence of liquidity constrained households and firms, and it raised poverty and inequality in the country. Figures 12a and 12b illustrate the rise in the Gini inequality index and the rise in the share of households at risk of poverty between 2009 and 2013. These developments further undermined the support for market-oriented reforms.
Figure 12a. The rise in inequality
Figure 12b. The rise in poverty
Source: ELLSTAT 2015
But then, what explains these wage and price dynamics, which at the same time depressed domestic and foreign demand?
The short answer is that, by focusing on labour rather than product market reforms, the Troika got the reform sequence wrong (see EC 2014). Here, labour market reforms are hailed as ‘major’ and ‘far reaching’, and they were indeed. They modified crucial features of the wage bargaining system, for example:
- Limiting the coverage and duration of collective agreements;
- Minimum wages were frozen;
- A system of arbitration was introduced; and
- Apprenticeship contracts were introduced.
Job protection was also reduced:
- Dismissal notice periods and severance pay were cut;
- Collective dismissals were made easier;
- Fixed term contracts were promoted, while at the same time duration limits for Temporary jobs were increased; and
- Working time became more flexible.
Much less enthusiastic is the narrative on product market reform. The EC reports that reforms were started in a number of fields such as barriers to entry in trade, retail distribution, professions; that red tape and tax costs were reduced and so on (EC 2014). However, it admits that despite progress being made in several areas, Greece still lagged behind countries at a similar level of development.
So what went wrong in Greece? Fundamentally, the economy eventually faced reality. The debt-financed growth bubble that occurred upon joining the euro eventually burst. The adjustment, however, was harsher and more painful than necessary. Fiscal consolidation was too tight and too front-loaded, and debt restructuring was too little and too late. The sequence of structural reforms was plainly wrong. By concentrating on the labour, rather than on the product market, the policies produced a sharp fall in nominal wages that was not paralleled by a similar reduction in prices. Domestic demand was depressed, and foreign demand did not pick up, while inequality soared.
This analysis has obvious implications for the ongoing negotiations between Greece and its lenders:
- The required fiscal adjustment should be smoothed over time, about 1.5% primary surplus would probably do (see Manasse 2015), and debt payment should be restructured accordingly;
- Structural reforms should concentrate on reducing barriers to entry in product and export markets, reducing bureaucracy, improving the tax administration and removing credit constraints for small and medium enterprises; and
- The reform of the pension system, while inevitable, should be implemented gradually with a view on not further depressing consumption by raising uncertainty on future incomes.
Athanasoglu, P P, C Backinezos and E A Georgiou (2010) “Export performance, competitiveness and commodity composition”, Bank of Greece Working Paper, no. 114.
EC (2014) “The second economic adjustment programme for Greece”, Fourth review, Directorate-General for Economic and Financial Affairs, April, European Economy Occasional Papers 192.
Giavazzi, F and M Pagano (1990) “Can severe fiscal contractions be expansionary? Tales of two small European countries”, NBER Macroeconomics Annual, 5: 75–111.
IMF (2014) “Greece 2013 Article IV consultation”, Country Report No. 13/154.
Manasse, P (2015) “Syriza and debt talks: Estimates from a Rubinstein bargain approach”, VoxEU.org, January 27.
1 The sharp fall in the most recent period in TFP is due to the fact that GDP growth fell much more rapidly than factors of production, resulting in a negative unexplained component (TFP).
2 see Giavazzi and Pagano (1990)
This article was first posted on voxeu.