On the debt and deficit criteria of the Maastricht Treaty

government debt

A lot has already been written and debated on the aims and (in)feasibility of having fiscal limits on excessive public borrowing. It is well understood that government debt, among other factors, is influenced by economic growth, periods of wars, demographic changes, financial crises, interest rate-growth differential, and economic cycles.

There have been many discussions and debates on the so-called “debt-growth controversy” following the (in)famous Reinhart and Rogoff’s (2010) paper and whether the causality goes from high debt to low economic growth or viceversa. The focus on debt issues partly started because, unlike the previous debt cycles that mainly happened during war times, since the mid 1970s almost all advanced economies experienced long-term increase in their public debt-to-GDP ratios that took place during peacetime.

This would then mean more sustained and structural needs for financing, including also the effects of one-off events as those of financial and banking crises and the Great Recession (e.g. in terms of post-crisis lost tax revenues). Also the by now famous concept of secular stagnation – a state of negligible or zero economic growth – if continued into the future, would imply a further decrease in governments’ capability of paying their debts.

The focus of this post is on the relationship between the two indicators of the fiscal criterion as defined in Article 126(2) of the Treaty on the Functioning of the European Union, according to which the Member States “shall avoid excessive government deficits” and comply with the “budgetary discipline” on the basis of the maximum reference values of the (planned or actual) government deficit to GDP (at market prices) ratio and government debt to GDP ratio.

The Protocol on the excessive deficit procedure of the Treaty specifies these reference values to be 3% and 60%, respectively. The figure above illustrates the Maastricht definition of the government debt ratio for 22 advanced countries for the year of 2016 (Source: OECD Economic Outlook 101 database). In particular, it clearly shows that 68% of the listed countries (i.e. 15 countries) do not comply with the Maastricht debt criterion.

The obvious question is, of course, why a 3% fiscal deficit-to-GDP ratio implies 60% government debt-to-GDP ratio threshold level (and/or vice versa)? For this purpose, one can readily use the Domar model developed by the Russian-American economist Evsey David Domar (1914-1997). For technical details, the reader is referred to Domar (1944, “The ‘burden of the debt’ and the national income”, American Economic Review, 34).

One of the conclusions of the Domar model is that if the deficit ratio, real growth rate and inflation rate are constant, then in the long run, the government debt-to-GDP ratio will converge, irrespective of its starting value, to the following limit: Debt limit = Fiscal deficit ratio / (Real growth rate + Inflation rate).

Debt limit 

This implies that the fathers of the Maastricht Treaty apparently presumed the long-term annual real GDP growth rate of 3% (a feasible reference value back then in the euro area) and the inflation rate of 2% (the target rate of the European Central Bank, ECB).

However, in retrospect we now know that the 3% real GDP growth rate turned out to be very optimistic, and also does not seem to be feasible in the coming future (e.g. recall the secular stagnation thesis). Instead, if we take a more realistic long-term economic growth rate of 1% and the ECB inflation target of 2%, then according to the Domar model, a 3% deficit ratio implies the debt limit of 100%.

If this would have been taken as a new debt criterion, then the list of “non-compliant” countries in the figure above exceeding this new value would dramatically reduce to only 4 countries (i.e. Belgium – 106% debt ratio, Portugal – 130%, Italy – 132.5%, and Greece – 179%). Thus, it can be noticed that such a change in debt limit would have had important (fiscal) consequences in abiding by the Stability and Growth Pact requirements for countries with the government debt ratio in-between 60% to 100% and the fiscal deficit ratio of up to 3%.

In reality, due to the diverse economic, social, political, and cultural factors, countries experience different economic growth and inflation rates. Let us see at such heterogeneity in more detail. Using the same OECD Economic Outlook 101 database, two separate time periods of 1992-2006 and 2012-2018 are considered, ignoring the years of and immediate to the Great Recession.

The average values of the real output growth and inflation rates over these time periods are given in the table below. Then in the subsequent two columns next to these figures the debt limits, as per Domar model, are presented that correspond to the average actual real growth and inflation rates. However, besides the Maastricht fiscal deficit-to-GDP ratio target, I also computed the debt limits that would correspond to the fiscal deficit of only 0.35% of GDP.

This figure is the Federal deficit limit as indicated in the 2009 amendment to the German constitution – “Basic Law” –  in order to introduce the concept of “debt brake” which allows for Federal fiscal deficit of only 0.35% of GDP as of 2016 and no deficits at all as of 2020 for the Federal states.

We do not go into the details of the table outcomes, one can her(him)self go through the implied debt limits and confirm that indeed there are a lot of debt limit differences not only among the listed countries, but also per country over the two considered time periods.

In the Euro area as a whole, for example, in the 1992-2006 period the average economic growth and inflation rates were roughly 2%, and thus the Maastricht-related (i.e. consistent with its 3% deficit ratio) debt limit would be 75%, which is 25% larger than the Maastricht 60% debt criterion.

On the other hand, in the 2012-2018 period, the Eurozone’s expected average real growth is roughly 1% and the expected inflation rate 1.15%, which together with the 3% deficit ratio would imply a debt limit of 140%. The last figure is larger than its Maastricht counterpart criterion by a factor of 2.33. (The computed negative debt limit for Greece reflects its 2012-2018 negative average economic growth and deflation rates, hence better be ignored as, in general, economic stability presupposes positive output growth and inflation rates.)

Finally, note that the debt ratio thresholds corresponding to the German Federal deficit limit of 0.35% are far much tighter than their Maastricht-related counterparts. In fact, per country the relevant debt limit makes only 11.7% of the debt limit corresponding to the 3% deficit ratio.

This can be also seen explicitly in the table along the row labelled “Ideal’ scenario” – “ideal” in the sense of using a more realistic for the EU countries real GDP growth rate and the ECB inflation target. All in all, one can conclude that in the practice of constraining excessive public borrowing, it would be fair and highly relevant to adequately account for the realities of different national economies, in particular, for their longer term expected economic growth and inflation rates prospective.

  1992-2006 2012-2018
Real GDP
(%)
GDP deflator
(%)
Debt limit (%) Real GDP
(%)
GDP deflator
(%)
Debt limit (%)
0.35%
deficit
3%
deficit
0.35%
deficit
3%
deficit
«Ideal» scenario 1 2 11.7 100 1 2 11.7 100
Argentina 8.69 10.94 2 15 0.69 26.95 1 11
Australia 3.36 3.66 5 43 2.68 0.83 10 85
Austria 2.32 1.78 9 73 1.12 1.80 12 103
Belgium 2.23 2.01 8 71 1.11 1.44 14 118
Brazil 3.51 9.02 3 24 0.05 7.25 5 41
Canada 2.84 2.80 6 53 2.05 1.30 10 90
Chile 5.74 7.48 3 23 2.78 3.46 6 48
China 10.82 4.63 2 19 7.08 2.04 4 33
Colombia 4.59 8.47 3 23 3.37 3.33 5 45
Costa Rica 4.82 12.70 2 17 4.00 3.96 4 38
Czech Republic 5.49 1.48 5 43 2.00 1.56 10 84
Denmark 2.35 2.08 8 68 1.34 1.18 14 119
Estonia 8.37 5.96 2 21 2.49 2.60 7 59
Finland 3.29 0.93 8 71 0.32 1.70 17 148
France 1.95 1.78 9 80 0.93 0.85 20 169
Germany 1.24 0.89 16 141 1.45 1.62 11 98
Greece 4.24 3.07 5 41 -0.99 -0.38 -26 -220
Hungary 3.95 6.44 3 29 2.39 2.53 7 61
Iceland 5.13 3.54 4 35 3.76 3.24 5 43
India 8.41 5.30 3 22 7.03 4.54 3 26
Indonesia 5.25 10.61 2 19 5.25 4.15 4 32
Ireland 6.02 2.75 4 34 6.59 1.42 4 37
Israel 3.94 0.64 8 65 3.28 2.00 7 57
Italy 1.29 2.54 9 78 -0.14 1.06 38 326
Japan 1.56 -1.03 67 571 1.19 0.51 20 176
Korea 4.68 2.26 5 43 2.79 1.48 8 70
Latvia 9.84 8.82 2 16 2.90 1.62 8 66
Lithuania 8.06 3.88 3 25 2.99 1.78 7 63
Luxembourg 3.64 3.82 5 40 3.74 1.34 7 59
Mexico 3.37 7.87 3 27 2.32 4.03 6 47
Netherlands 2.22 2.09 8 70 1.26 0.98 16 134
New Zealand 3.64 2.25 6 51 2.96 1.67 8 65
Norway 2.65 5.90 4 35 1.59 1.32 12 103
Poland 4.59 5.03 4 31 2.79 1.05 9 78
Portugal 1.16 3.33 8 67 0.34 1.27 22 186
Russia 7.24 17.19 1 12 0.77 6.68 5 40
Slovak Republic 5.96 4.12 3 30 2.89 0.56 10 87
Slovenia 4.21 3.18 5 41 1.58 1.25 12 106
South Africa 4.24 6.64 3 28 1.43 5.63 5 42
Spain 3.48 3.91 5 41 1.20 0.54 20 172
Sweden 3.37 1.34 7 64 2.26 1.66 9 77
Switzerland 2.29 0.91 11 94 1.48 -0.15 26 225
Turkey 6.88 24.01 1 10 4.93 7.65 3 24
United Kingdom 2.89 2.48 7 56 1.84 1.64 10 86
United States 3.19 2.58 6 52 2.14 1.71 9 78
Euro area 1.99 2.01 9 75 0.99 1.15 16 140
Total OECD 2.82 2.80 6 53 1.87 1.69 10 84
Minimum     1.1 9.7     -26 -220
Maximum     67 571     38 326
Source: OECD Economic Outlook 101 database and own calculations.      
Umed Temurshoev

Umed Temurshoev

Profesor adjunto en la Universidad Loyola Andalucía. Licenciado en Economía por la Universidad Americana de Asia Central (2003, Kirguizstán), obtuvo su Máster en Economía por el Centro de Investigación Económica y Educación de Postgrado – Instituto de Economía (2005, República Checa) y es Doctor en Economía por la Universidad de Groningen (2010, Países Bajos). Fue investigador postdoctoral y profesor en la Universidad de Groningen entre 2009 y 2012. Posteriormente, ha trabajado como investigador para el JCR-IPTS de la Comisión Europea durante el periodo de 2012-2015. Dos de sus artículos fueron galardonados con los Leontief Memorial Prize (2012) y Lawrence R. Klein Award (2014, junto con el Profesor Erik Dietzenbacher). Sus líneas de investigación son la modelización económica, el análisis input-output, el análisis bayesiano aplicado y la programación matemática.

Deja un comentario

Tu dirección de correo electrónico no será publicada. Los campos obligatorios están marcados con *