Empirical work on debt cycles and debt sustainability has been constrained by lack of public debt data on a large number of countries over a long time period. Existing studies are based on datasets that either cover short time periods (such as Jaimovich and Panizza 2010) or omit a large number of countries (such as Reinhart and Rogoff 2010). In our latest study (Abbas et al 2011), we compile a comprehensive historical public debt database covering 178 countries, starting from 1880 for G7 countries and a few other advanced and emerging economies, and from 1920 for additional advanced and emerging economies. For low-income countries, data coverage generally starts in 1970 (Abbas et al 2011).
Figure 1. Debt-to-GDP ratios across country groups, 1880–2009 (Group PPPGDP-weighted averages, in percent of GDP)
Figure 1 provides a broad historical perspective of debt developments in advanced, emerging, and low-income economies. Debt levels in advanced economies (now the G20) averaged 55% of GDP over 1880–2009, with a number of peaks and troughs that correspond with key historical events along the way.
- During the first era of financial globalisation (1880–1913), debt ratios trended down as public finances were, for the most part, under control, and growth was supported by an unprecedented level of gold standard-enabled financial and trade flows. Debt ratios reached their lowest level – 23% of GDP in advanced economies – in 1914, when the First World War began.
- But war and the fiscal crises that followed, the Great Depression (early 1930s), and World War II (1941–45) drove debts upward (to almost 150% of GDP in 1946).
- By 1960, however, debt ratios had declined to 50% of GDP on the back of strong postwar reconstruction and in some cases moderate, high, or even hyper inflation.
- Debt ratios began to rise again starting in the mid-1970s, with the end of the Bretton Woods system of exchange rates and the two oil price shocks. Expanding welfare states, moderating growth, and higher interest rates all contributed to this seminal peacetime increase, which the present crisis has exacerbated.
Episodes of large debt reductions and build-ups in advanced economies
So much for aggregate trends; what about individual episodes? For a group of 19 advanced economies, we identify 68 debt declines (including seven defaults) and 60 debt increases sized greater than 10% of GDP (see Figures 2 and 3).1 The ‘non-default’ debt declines averaged 38% of GDP, and were distributed roughly evenly across four periods: the pre-1914 gold standard era, the two World Wars and intervening decades, the Bretton Woods years from 1946–70, and the post-1970 period. Debt surges averaged 44% of GDP and were bunched around 1914–45 and the peacetime period of 1970–2007.
Figure 2. Identified episodes of debt-to-GDP Decreases (in percent of GDP)
Figure 3. Identified episodes of debt-to-GDP increases (in percent of GDP)
Drivers of large public-debt changes in advanced economies
So what drove these large debt movements? To find out, we decompose changes in the debt-to-GDP ratio into cumulative contributions from the primary deficit, the interest-growth differential (often referred to as the ‘automatic debt dynamics’), and a stock-flow adjustment residual which could reflect a number of factors: currency valuation effects operating on foreign currency debt; assumption of debts of non-governmental entities; debt restructuring or default; privatisation or drawdown and build-up of government deposits (see Escolano 2010).
Table 1 summarises the decomposition results for large debt increases. While wartime debt increases started from higher debt levels and were associated with larger primary deficits, they were smaller in size than peacetime surges. The key driver of this appears to be the interest-growth differential component, which was relatively modest during wartime (2% of GDP), but sizable during peacetime recessions (20% of GDP, prior to 2007, and 10% of GDP during the present slowdown). Also, debt increases during non-recessionary periods were slightly larger than those during global recessions despite the former commanding debt-reducing negative interest-growth differentials. The key contributors to these ‘good time’ debt surges were runaway primary deficits and, especially, stock-flow adjustments which averaged 34% of GDP during (20% of GDP over all 60 debt increases during good and bad times). It appears that these stock-flow adjustments reflect governments absorbing liabilities ‘below the line’, while shielding headline fiscal balances ‘above the line’, which is consistent with the pattern documented by the IMF (2011a and 2011b) for more recent periods.
Table 1. Decomposition of 60 large debt ratio increases for different sub-samples (averages, in percent of GDP)
Notes: 1/ Periods of war relate to the World Wars (1914–18, 1939–45); and other individual wars occurring outside these two periods (see paper for details). 2/ Recessionary periods include the Long Depression which started in 1873 (effectively 1880 in our sample), and lasted until 1896; the post-WWI recession in Europe and the North America (1919–21); the Great Depression (1929–32); the 1973–75 recession following the oil price shock and the collapse of the Bretton Woods system of exchange rates; and the 1980–82 and 1990–92 recessions. 3/ For the Great Recession, debt increases are computed over the 2007–13 period, and the decomposition thereof is based on the IMF April 2011 WEO projections. These 17 episodes (19 countries in our sample less the two countries, Sweden and Switzerland, whose debts were projected to fall) are not part of the 60 episodes otherwise identified over the period 1880–2007.
Table 2 illustrates the components of large debt decreases across various sets of circumstances – starting points, time periods, and duration, pace, and magnitude of debt reduction. Debt declines featuring a default averaged about 64 percentage points of GDP, much larger than non-default declines, and were reflected in large stock-flow adjustments (37 percentage points). The 61 non-default episodes registered an average reduction of 38% of GDP, and were accounted for by the primary balance and the growth-interest differential components in roughly equal amounts.
Interestingly, the post-World War II non-default debt reductions, aligned with the Bretton Woods years (1945–70), were associated with a dominant contribution from the growth-interest differential component: real growth averaged around 4%-5% annually in these episodes while the average real interest rates paid on debt were negative. These findings tend to support recent conclusions by Reinhart and Sbrancia (2011) that financial repression policies had a role to play during this period, characterised by capital controls, in reducing debt burdens in advanced economies.
Table 2. Decomposition of 68 large debt ratio reductions over different sub-samples (averages, in percent of GDP)
Notes: 1/ Includes episodes which either overlapped or immediately followed hard or soft defaults (the latter include forced currency conversions) on central government domestic and/or external debt (as documented in Reinhart and Rogoff 2009).
Implications for current debt debate
The composition of the 11 debt reductions observed during 1880–1914, the first era of financial globalisation, is quite similar to that witnessed in the post-1970 financially liberalised period. In both cases, the debt ratio reductions were mainly caused by large primary surpluses. In fact, the post-1970s debt reductions are accounted for almost entirely by primary surplus improvements. However, insofar as such improvements are boosted by the cycle and easier to implement in the context of strong growth, these results may somewhat understate the true role of growth in debt declines; strong growth was a consistent feature of most debt decline episodes.2 That conventional fiscal adjustment and growth have led the way in periods of global financial integration is intuitive as well as consistent with previous studies (such as IMF 2010).
Looking ahead, highly indebted advanced economies are confronted by a challenging landscape. The pursuit of unconventional options – such as reverting to financial repression policies akin to those taken during the post-WWII years, reducing the burden of domestic debt through higher inflation, or restructuring – may be a tempting shortcut but it comes with high costs. A gradual but steady adjustment is the right way to go. History shows an orderly adjustment is much easier in the context of sustained medium-term growth. This suggests that there is a premium on both implementing structural measures that improve competitiveness and the business environment, and designing fiscal adjustment in a manner that minimises the drag on growth.
Abbas, SM Ali, Nazim Belhocine, Asmaa El-Ganainy, and Mark Horton (2011), “Historical Patterns and Dynamics of Public Debt – Evidence From a New Database”, IMF Economic Review, 59(4):717-742.
Escolano, J (2010), “A Practical Guide to Public Debt Dynamics, Fiscal Sustainability, and Cyclical Adjustment of Budgetary Aggregates”, Technical Notes and Manuals, Fiscal Affairs Department, International Monetary Fund.
International Monetary Fund (2010), World Economic Outlook, April 2010.
International Monetary Fund (2011a), The Fiscal Monitor, IMF Fiscal Affairs Department, April.
International Monetary Fund (2011b), The Fiscal Monitor, IMF Fiscal Affairs Department, September.
Jaimovich, D and U Panizza (2010), “Public Debt Around the World: A New Data Set of Central Government Debt”, Applied Economics Letters, 17(1):19-24.
Reinhart, Carmen, and Kenneth Rogoff (2010), “This Time is Different Chartbook: Country Histories on Debt, Default, and Financial Crises”, NBER Working Paper No. 15815.
Reinhart, Carmen, and Kenneth Rogoff (2011), “A Decade of Debt”, NBER Working Paper No. 16827.
Reinhart, Carmen, and M Belen Sbrancia (2011), “The Liquidation of Government Debt”, NBER Working Paper No. 16893.
1 The countries covered are: Australia, Austria, Belgium, Canada, Finland, France, Germany, Greece, Ireland, Italy, Japan, Netherlands, New Zealand, Portugal, Spain, Sweden, Switzerland, the United Kingdom, and the United States.
2 Interestingly, the role of the growth-interest differential varied across the two periods, mainly on account of the interest component, which was much higher in the post-1970 setting than the pre-1914 one.