For several years now, the Eurozone has been engaged in a long period of austerity in order to avoid sinking further into crisis. The aim of this type of policy, which consists of cutting government spending while trying to increase revenue, was to prevent countries’ public debt from exploding and becoming unsustainable (1). Excessively high levels of public debt can prove devastating when they continue to rise, raising the prospect of future difficulties for the indebted government in repaying it. This creates a vicious circle: the stock of debt increases, interest rates (2) rise, leading to higher interest payments that exacerbate the public deficit, and the deterioration of the latter requires new bond issues and/or loans at high interest rates. To break this negative spiral, the path taken was to rebalance public accounts and reduce public debt.
However, in times of crisis, reducing public spending at a time when private investment is drying up can be dangerous for economic activity. If the decline in spending amplifies the slowdown in activity, its effects quickly become undesirable and risk plunging countries into a longer and harder recession. Beyond the fact that entering a cycle of debt reduction sends a strong message to the markets and reassures them (3),the impact on the real economy remains highly uncertain. Before embarking on this type of policy, it was necessary to measure the « costs and benefits » of an austerity policy and see whether it would be able to deliver satisfactory results to meet current expectations. It was then established that the success of this type of policy was conditional on the fulfillment of the following two assumptions: a low fiscal multiplier and Barro-Ricardian equivalence.
Fiscal multiplier: explanations, estimates, errors
First, the fiscal multiplier is a measure of the impact of a change in public spending on GDP. It is generally agreed that, all other things being equal, a decrease in spending generates a certain contraction in GDP. In the case of an austerity policy, the impact of the reduction in spending must have as little effect as possible on the change in GDP, so as not to plunge the economy into recession; if this is the case, the fiscal multiplier is low. For example, with an estimated multiplier of 0.5, a €1 contraction in public spending leads to a 50-cent decline in GDP. The fiscal multiplier is calculated empirically for each country, based on historical data on several variables for a given period. It is based on a key assumption: the absence of Barro-Ricardian equivalence. The latter is supposed to reflect the behavior of economic agents (households and businesses) in response to a change in public spending. In a context of economic slowdown, an increase in public spending does not have the desired stimulus effects, as agents will tend to save more rather than consume. This behavior is justified either by the fact that they anticipate a future increase in taxes (in response to the current increase in public spending) and therefore prefer to build up « reserves, » or because in times of crisis there is a high degree of uncertainty and it is better to build up savings to avoid finding oneself in difficulty later (due to unemployment, loss of purchasing power, etc.) (4). On the other hand, when fiscal policy is restrictive, the opposite behavior is observed, with economic agents drawing on their savings to consume and thus maintain a certain standard of living. Therefore, if this hypothesis is verified, reducing spending implies a fairly low fiscal multiplier, given that economic activity benefits from increased consumption and investment by so-called « Ricardian » agents.
The International Monetary Fund (IMF), the Organization for Economic Cooperation and Development (OECD), and the European Commission (EC) initially estimated that the fiscal multiplier averaged around 0.5 in the eurozone. This figure was the consensus level for 21 so-called advanced economies over the period 1970-2007. However, the gap between growth forecasts for Eurozone countries and the actual level was quite significant, with several countries affected by recession, leading these institutions to revise their calculation of the multiplier. Unfortunately, since 2007, economic conditions have changed andthe fiscal multiplier has been overall underestimated by a magnitude of 1 to 1.7 (5). A recent IMF publication from June 2012 shows empirically (6) that, in times of crisis, the increase in consumption and production is twice as large as the increase in public spending. It therefore appears that the basic assumptions on which the multiplier calculation is based have not been validated. We will now try to understand the reasons for this, using the example of Spain, but we will not attempt to calculate its level ourselves.
Multiplier and the Spanish economy
A model student in terms of public finance management during the 2000s, Spain has experienced numerous difficulties since 2007 and is now in the throes of a serious economic crisis: in 2012, it recorded a recession of 1.45%, a record unemployment rate in the eurozone of nearly 24.8%, and a deficit of 8.4% of GDP. The deterioration in public finances was due in particular to the massive increase in public spending to bail out the banks following the real estate crisis and to support regions whose deficits had increased considerably during the 2000s. Public debt rose from 36.3% of GDP in 2007 to 86% in 2012. Numerous reforms (increase in VAT, reduction in public spending, freeze on the thirteenth month’s salary for civil servants, labor market reform) were undertaken to limit the increase in public debt, and Spain, like the rest of its Eurozone neighbors, entered a phase of austerity. As mentioned above, this cure was necessary and was validated, among other things, by forecasts of a low multiplier and the likely « Ricardian » behavior of economic agents.
Firstly, households: these tended to take on a lot of debt between 2000 and 2006, with a significant deterioration in their solvency ratio. Following the real estate boom and the ensuing banking crisis, unemployment rose sharply (7) and households, fearing they would lose their jobs in the face of a catastrophic economic situation, built up savings. Between 2006 and 2009, household savings doubled and the savings rate rose from 10.25% to 17.8%. At the same time, the Spanish government significantly increased its spending to partially replace private investment and support economic activity as best it could. This reaction by households, combined with expansionary fiscal policy, suggests that at the time we were indeed seeing « Ricardian » behavior at the onset of the crisis. From 2010 onwards, faced with deteriorating public finances and the significant risk of contagion in the eurozone with the Greek crisis, Spain entered a phase of fiscal austerity. It was also from 2010 onwards that private consumption began to recover, reaching 59.3% of GDP in 2012 (compared with an average of 57.1% since 2005), thanks in particular to households drawing on their reserves and increasing their consumption. Here again, households adopted « Ricardian » behavior, sacrificing their savings to consume more during a period of declining public spending. Based on these factors, the conclusion is that the Barro-Ricardian equivalence hypothesis is verified. Therefore, choosing a low multiplier, not necessarily equal to 0.4 but less than 1, seems appropriate.
However, given the negligible growth results, there is room for doubt about the real impact of austerity on GDP variation, which is much higher than this figure, reaching 1.7 according to some economic research departments. A study of Spanish businesses and other non-financial companies may provide some answers. Unlike households, businesses have tended to be « non-Ricardian » since the decline in public spending began. Corporate savings have increased considerably, and this has been amplified by the crisis. When private investment fell, public investment attempted to take over, to a lesser extent, without success until the first reforms arrived and the government found itself with its hands tied and unable to continue increasing its deficit, or at least to make cuts in its spending. Companies had no other means of self-financing than by accumulating greater savings (8). In 2006, the self-financing rate (savings on investment) was 35%, but reached 110% in 2011. These savings enabled companies to generate positive financing capacity, which they have largely devoted to reducing their debt since 2009. These savings have undoubtedly also helped to keep their margins at a reasonable level, but this is still insufficient to influence a return to investment. Investment (9) seems vital to supporting economic activity and is weighing on GDP growth. This « non-Ricardian » attitude of Spanish non-financial companies is consistent with a high fiscal multiplier.
The rush towards austerity among Eurozone member countries appears to be another factor that may have contributed to an inaccurate estimate of the fiscal multiplier, due to its effects on foreign trade. The main trading partners of Eurozone countries are essentially other members of the zone. The share of exports from one Eurozone country to others has varied between 40% and 65% over the last decade. When all countries simultaneously reduce their spending, their trade declines and each country finds itself constrained in terms of external markets. This symmetry is highly disadvantageous and is generally accompanied by a general deterioration in each country’s trade balance (exports minus imports), thereby affecting their GDP. In the case of Spain, its main partners are France, Italy, and Portugal (which have accounted for an average of 37% of total exports since 2000), three countries that have also tended to pursue austerity policies on the one hand and import less on the other. This case is somewhat unusual, as Spain has managed to offset the decline in its exports to the Eurozone by exporting more to the rest of the world, especially to South America. When a country does not diversify its export regions sufficiently, this alternative does not exist and the trade balance deteriorates all the more. The inability to benefit from external demand from its main trading partners, constrained by restrictive fiscal policies, undoubtedly plays a role in the contraction of GDP and should therefore be included in the calculation of the fiscal multiplier. The symmetrical aspect of the simultaneous adoption of fiscal austerity measures automatically increases the level of the multiplier.
Conclusion
The fiscal multiplier in the eurozone is closer to 1 than 0.5 and would be even higher for several countries, particularly those on the periphery, including Spain. The IMF has conceded that it made a mistake and that this will be corrected, as it now considers the multiplier to be close to unity. In Spain, two factors explain the inaccurate forecast of the fiscal multiplier: the non-Ricardian behavior of Spanish companies and the deterioration of intra-zone trade with other eurozone countries. The first had the opposite effect to that hoped for in the context of restrictive fiscal policy. The second was not taken into account in the calculation of the Spanish fiscal multiplier. It would appear that, after embarking on a race to drastically reduce public spending, the eurozone countries are now preparing to simultaneously step up their efforts to quickly get back on the path to competitiveness. Let us hope that this time, the potential gains and losses of future « competitiveness policies » have been carefully weighed up in advance.
Notes:
1 – According to the IMF, a country’s debt is considered unsustainable based on three criteria: the trajectory of the debt-to-GDP ratio, the stabilization of the debt-to-GDP ratio at a level that allows for the refinancing of public debt while preserving economic growth, and the composition of the debt, which determines the risk of a negative spiral in the event of a deterioration in public finances.
2 – For a range of bonds with maturities or maturities spread over time.
3 – And therefore to benefit from lower interest rates or strong demand for bond issues, two factors that facilitate the financing of public debt on the financial markets.
4 – Even if this second argument is not mentioned in the Barro-Ricardian equivalence concept and is more a matter of intertemporal smoothing of income and consumption, it nevertheless has the advantage of being consistent with the theory while complementing it to better reflect reality.
5 – For the IMF, it is greater than 1, but several studies go further, claiming that it could be closer to 2 than 1.
6 – Corsetti, Meier & Mueller « What determines government spending multipliers, » IMF Working Paper, 12/150, June 2012.
7 – As the construction sector employed a large number of workers before the crisis, its collapse had a very negative impact on employment.
8 – The distribution of companies’ added value benefited savings rather than the wage bill, which fell significantly overall due to the non-renewal of temporary contracts, which are very common in Spain.
9 – Investment contributed only 21% to GDP in the fall of 2012, according to the EC.
References:
IMF, « World Economic Outlook: Financial stress, downturns, and Recoveries, » 2008, page 177.
Carlo Cottarelli, Reza Moghadam, « Modernizing the Framework for Fiscal Policy and Public Debt Sustainability Analysis, » IMF, August 5, 2011.
Giancarlo Corsetti, André Meier, Gernot Müller, « What Determines Government Spending Multipliers? », IMF, June 1, 2012.
European Commission, European Economic Forecast, Autumn 2012, page 69.
