photo: Jeff Golden
Anxious about the threats to Europe from Syria, Turkey, the refugee crisis, terrorism, Brexit? Be patient. Europe is now at the autumn war of decimals. The governments of the member countries of the Euro Zone are submitting to the Commission the three-year budgetary plans in compliance with the fiscal regulations of the Treaties. The ordinary European citizen may have learned that his/her government should not exceed a deficit/GDP ratio of 3% per year. Hence he/she may feel comfortable with official data (Eurostat) saying that the EZ is expected to stop at 1.7% in 2016 and 1.5% in 2017. Only France and Spain will exceed the threshold. Indeed, under the austerity therapy, deficit/GDP ratios have been consistently declining throughout the EZ from the peaks reached in the aftermath of the Great Recession. So the ordinary European citizen (at least outside France and Spain) may think that no further austerity will be prescribed by the Commission, or even that some margin exists for a fiscal boost to a stagnating economy with persistent unemployment and decline in workers’ incomes. These feelings of the ordinary European citizen may be corroborated by a study of the OECD (2016) showing that almost all the EZ countries in fact have fiscal space (i.e. possibility to extend fiscal stimuli while keeping public debt sustainable) that can be targeted “to escape from the low growth trap”. Nonetheless, the Commission’s view is that
“The fiscal requirements contained in the country-specific recommendations of the Council would lead […] to a moderately restrictive fiscal stance for the euro area as a whole in 2017 and 2018, while the economic situation would seem to call for an expansionary fiscal stance” (Communication 727).
The ordinary European citizen may be astonished by this symptom of schizophrenia, possibly because he/she has missed one remarkable innovation introduced in the fiscal regulations of the EZ during the 2010-12 crisis: the Medium Term Objective (MTO) of budget plans. Its aim is to commit governments to keeping the so-called structural budget on track towards the target of “zero or small surplus”. In practice, as long as the MTO is not met, the government faces a stricter limit on the budget than the 3% deficit/GDP ratio. This is for instance the case of Italy, which in spite of being well below the 3% deficit ceiling, results to have a structural deficit that ought to be cut (see table 1).
Unfortunately, the structural budget is quite a tricky entity: it is neither observable in the public accounts nor is it easily calculable. On paper, the structural budget (SB) is an accounting residual after subtracting from the total budget (TB) its cyclical components (CC), i.e. revenues and expenditures automatically activated by the business cycle, and the one-off measures (OOM) taken by the government for various extraordinary and temporary contingencies. That is to say,
SB = TB – CC – OOM
On the one hand, the SB acknowledges that the budget may deviate from its MTO owing to the ups and downs of the economy (measured by CC) or to temporary events and decisions (accounted for by the OOM items). On the other hand, the Commission wants to be sure that the required budget adjustments are permanent rather than due to temporary “window dressing” (also accounted for by OOM). But obtaining these figures is highly contentious.
With some pain, I have sought to reconstruct intelligibly the relevant figures for Italy in 2016 and 2017 based on Eurostat figures (table 1).
Table 1. Italy’s structural budget 2016-17 (- indicates deficit)
Source: Elaborations on Eurostat, Database AMECO
Classifying which entries of the budget are one-off is controversial. The largest part of the public budget results from laws or decrees. Some may explicitly contain a deadline; more often they do not, and the government may later decide to extend them or not. As can be seen in the case of Italy, in 2016 a €2.2 billion surplus results as OOM and is subtracted from the budget widening the SB deficit; for 2017, this figure escalates to €4.5. Italy’s claim that the expenditures for refugees and the earthquakes should be recorded as OOM would reduce its amount and hence the SB deficit. Yet these expenditures are certainly extraordinary, but also not so temporary, and so on so forth.
However, the truly severe problems arise with the cyclical components. These are not readable in the public accounts, and are the result of complex econometric estimates of two indicators. First, the cyclical component of the observed GDP has to be identified. Then, regressions of revenues and expenditures on this indicator yield the estimate of the (semi)elasticity of the budget to the cycle. Thus, when the business cycle is negative the Commission can estimate the deterioration of the budget (CC < 0) due to its cyclical components (see e.g. Mourre et al. 2013)
Each step of this procedure is fraught with unsettled problems as testified by academic research (e.g. Cottarelli 2015, Fioramanti and Waldmann 2016). First, the Commission’s econometric model is based on a particular theoretical definition and measure of business cycles: namely the percent gap between actual GDP and its potential level. The underlying theory is the New Keynesian version of dynamic stochastic general equilibrium models (DSGE) where potential GDP is given by the evolution of full-employment supply-side factors (technology and endowment of production factors), and actual GDP may fluctuate around its potential owing to demand shocks plus “frictions” (typically wage and price stickiness). Other measures would be possible, such as the simple year rate of change of GDP, or its deviation from a historical trend. Why has the Commission adopted this particular theory?
One consequence is the necessity to estimate the potential GDP. Various models and techniques have been developed, none of which dominates the others. Notoriously, different official agencies produce different estimates of potential GDP and consequently of output gaps: see figure 1 for the comparison between Eurostat and OECD. After the Great Recession, Eurostat has systematically estimated better gaps than the OECD, and hence lower CCs and larger structural deficits. Differences are so substantial at the country level that the verdicts of the MTO assessment can be overturned (Fioramanti and Waldmann 2016).
Figure 1. Estimated output gaps for the Euro Zone
Source. Eurostat, AMECO Database; OECD World Economic Outlook Database
Without probing here into the technicalities, it is worth recalling that one general problem of the estimates of potential GDP is that it results correlated with the actual GDP. This problem seems particularly acute with the Eurostat estimates: the correlation coefficients range from 0.80 (Portugal) to 0.98 (Austria and Belgium), with the notably low outlier of 0.64 for Italy. Positive correlation implies that when say the actual GDP falls, the estimated potential GDP falls, too, to the effect that the output gap results small. Italy is an emblematic case as shown by figure 2.
Figure 2. Eurostat values of Italy’s GDP and potential GDP at constant prices (2010)
Now, two possible interpretations arise. The first is that the underlying theory is misleading in that the neat separation between GDP fluctuations and its potential trend ( or between aggregate-demand and aggregate-supply shocks) does not hold in reality. Several explanations of aggregate demand-supply interactions are available, the most famous one being hysteresis. The second is that the estimation technique of potential GDP is flawed. The bottom line is that serious policy mistakes are induced in either case.
Turning to the estimation of the budget elasticity to the cycle, again one finds a bewildering variety of techniques and results. To mention just one problem, the legislative production underlying the extension and operation of the so-called automatic stabilisers is quite different across countries. Furthermore, each country may modify its own stabilisation mechanisms not so unfrequently, so that the degree of structural stability necessary to obtain reliable estimates of the relevant parameters may be a chimera.
The MTO apparatus is also flawed conceptually. In the New Keynesian DSGE framework, the potential GDP is an equilibrium concept, such that no policy action is necessary. In particular, the public sector should be in balance. A structural deficit relative to potential GDP is therefore an indicator of the budget adjustment consistent with the return of the public sector to equilibrium. But what is the normative meaning of the public sector’s structural deficit when the economy is not in equilibrium at its potential GDP? It is reasonable to believe that if the economy were in equilibrium at potential GDP all entries in the public budget would be different, not only the cyclical ones. And even conceding that the Commission could exactly estimate the SB that would be generated by the economy being in equilibrium, what is the rationale of having governments to adjust, by means of true euros of the taxpayers, a hypothetical budget relative to a hypothetical state of the economy? It is likely that this exercise in hypothetical fiscal policy ends up in a real fiscal restriction which may widen the distance between the real and the hypothetical state of the economy, as candidly admitted by the Commission in its communication.
The problems discussed here are far from being merely technical. If the Stability and Growth Pact has been debated ever since its inception, it is time to recognise that the whole set of reforms introduced during the crisis (“Six Pack”, “Two Pack”, “Fiscal Compact”, etc.) has failed on all the requirements of good regulations: legitimacy, efficacy, simplicity, transparency. The result is a poisonous mixture which is intoxicating the relationships of governments with institutions, of governments with governments, and of peoples with peoples, to the advantage of the anti-European forces.
Cottarelli C. (2015), “Potential Growth Rates and the Working of SGP Fiscal
Rules”, Vox-EU, 2 March.
Fioramanti M., Waldamnn R. (2016), “The Stability and Growth Pact:
Econometrics and Its Consequences for Human Beings”, Vox-EU, 19 November.
Mourre G., Isbasiou G-M., Paternoster D., Salto M. (2013), “The Cyclically
Adjusted Budget Balance Used in the EU Fiscal Framework: An
Update”, European Economy, Economic Papers, n. 478.
OECD (2016), Using Fiscal Levers to Escape the Low Growth Trap,