Should we no longer worry about the level of debt?

In October 2008, at the peak of the Great Recession, the International Monetary Fund (IMF) confirmed that a reduction of debt at a global level was necessary and inevitable. Inevitable? Recent figures from McKinsey (2015) indicate that the debt levels of most countries in 2014 actually exceeded those of 2007: global debt has gone from representing 269% of GDP to 286%.1 And if the timeframe is widened, the data suggest that this upward trend in debt has been going on for a long time: according to the Bank for International Settlements (BIS), in the three decades between 1980 and 2010 the most developed countries increased their debt at an average annual rate equivalent to 5% of GDP. And in those countries for which we have a longer historical record, such as the US, a similar trend can be seen in private debt at least since the end of the Second World War. Given this situation, should we be concerned about this secular pattern of greater debt? Because, although the IMF was mistaken to think that deleveraging was inevitable, perhaps it was right to deem it necessary.

To answer this question, first we need to avoid demonising debt per se. Debt is a financial instrument related to a fundamental mechanism to create well-being: credit. It fulfils vital functions to ensure modern economies function well: it allows individuals and firms to temporarily adjust their consumption and investment, helps the public sector absorb shocks affecting economies and enables spending capacity to be spread across generations, as well as resulting in the more efficient allocation of capital for different uses in the economy and more efficient assignment of risk, discriminating between those taking on debt who are most able to assume it.

Nonetheless, not demonising debt is different from throwing all caution to the wind when accumulating debt in the long term. In economic terms we should be concerned about the sustainable rate of growth. To examine this question further, we need to make a fundamental distinction between a debt ratio that is compatible with normal growth in the economy and a ratio that creates excessive liquidity, excessive being more liquidity that the level required by the economy. Unlike in the sustainable situation, under certain conditions the undesirable outcome of a predominance of excessive debt accumulation could be unwarranted increases in the price of certain assets (such as harmful financial bubbles).2 In spite of its importance, in practice it is no easy task to distinguish between a sustainable and excessive rate of growth in debt. Conceptually the long-term factors determining the process of indebtedness differ in nature. The trend in debt critically depends on demographics, economic development, the degree of global economic and financial integration, the quality of institutions, financial innovations (in particular those allowing risk to be shared) and the overall stability of the macroeconomic framework. Although it is difficult to determine cause and effect, it is generally accepted that a positive trend in these variables is accompanied by a long-term trend of increasing debt.3 As the BIS pointed out: «Without finance and without debt, countries are poor and stay poor». But with too much debt perhaps they begin to be less rich.

Intuitively we can accept that there must be a limit as from which debt no longer strictly fulfils its functions and starts to generate risks and costs. The fundamental risk is that, in a situation of excessive debt, a debtor may be more vulnerable to possible shocks in terms of income, interest rate or uncertainty. An unexpected change in the flow of predicted income, worsening of financing conditions or deterioration in expectations represent a greater risk in a scenario of excessive debt. But even in the absence of such shocks, excessive debt entails costs. For example, there is evidence that companies with too much debt can make less than optimal decisions regarding corporate investment, rejecting profitable projects whose returns would be allocated to serve debt taken out in the past. When such behaviour spreads, the aggregate outcome is less growth. Richard Koo analysed how some Japanese firms went from being maximisers of profit to minimisers of debt and worsened Japan's long recession in the 1990s as they helped to make expansionary monetary policy less effective. Moreover, if the dreaded shock actually occurs, the financial sector might start to limit credit.

Within this general framework, one prominent stream in economics literature has tended to tackle the issue of excessive debt with an aggregate, empirical focus, aiming to determine the debt threshold as from which negative effects occur in economic growth. One of the most widely commented lines of research in the last few years in this area has been the work by Reinhart and Rogoff.4 According to the authors' estimates, those countries above the public debt threshold of 90% of GDP tend to record growth that is slightly lower than countries below this threshold. Although these figure were subsequently questioned due to errors of calculation (errors which, once corrected, concluded that the threshold in question did not appear), it is true that other authors, using different data and methodologies, still find that thresholds exist.5

This line of research can be complemented in at least three directions. The first is the so-called «balance sheet approach». The basic idea underlying this approach is that the capacity to repay debt depends not only on the level of debt but also on its structure (paying particular attention to aspects such as its duration or currency) and the quality of the assets that must generate the capacity to repay the debt or, in short, the relation between assets, debt and equity. This approach, initially designed to explain solvency and credit risk from a company point of view, has proven to be fruitful and has been extended to more aggregate levels. For example, the IMF has developed this approach to improve explanations of why emerging economies have historically had problems in repaying their debts even though these are at relatively low levels. A second approach attempts to find alternatives in order to establish the threshold. One significant example of this approach is a recent study by the IMF (2014) that uses empirical evidence based on historical data from a large number of countries, concluding that it is the trend in debt rather than any hypothetical threshold that conditions the rate of growth.6 In particular, the authors claim that the growth rate is similar for both countries with low levels of debt and those with higher levels but a downward trend. Lastly, a third variant revises the issue of thresholds by broadening the type of debtor: instead of analysing only public debt, as Reinhart and Rogoff did, it also includes companies and households. This is a useful adjustment for those who agree that the threshold concept should not be abandoned but that greater precision is required to quantify it, arguing that patterns of debt and deleveraging are not uniform between economic agents. One illustrative study from this point of view, carried out by BIS economists (2011) on developed countries and over a timeframe of three decades (1980-2010), has discovered two thresholds: 85% of GDP in the case of public debt and households and 90% for corporate debt. The direct conclusion from these thresholds is that the world economy is significantly at risk due to excessive debt. According to the McKinsey data, in 2014 Ireland exceeded the thresholds in all three sectors while Japan, Singapore, Belgium, the Netherlands, Spain, Denmark and the United Kingdom recorded excessive debt in two of the three types of debtor.

In short, the question of debt, always fundamental, continues to pose considerable challenges of interpretation for economists and, by extension, for society. The aforementioned approaches provide an example of the renewed attention being paid to determining what constitutes an excessive level of debt, a dangerous trend in debt accumulation or the quality of assets as a conditioning factor for debt sustainability. However, the challenges do not end there. From a more general point of view, the limitation stated by Olivier Blanchard in 2012 needs to be overcome (namely that much of modern macroeconomics «assumed that we could ignore much of the details of the financial system»), as well as reclaiming the importance given by authors considered canonical in their day (Wicksell, Hayek and Minsky, to cite three obligatory references) to the links between banking, credit or debt and well-being.

Àlex Ruiz

Macroeconomics Unit, Strategic Planning and Research Department, CaixaBank

1. Total debt of households, non-financial firms and general government (not consolidated).

2. Debt is sustainable when its accumulation and use do not generate an excessive burden in relation to future income.

3. Although the text focuses on the link between increasing debt and long-term variables, these also influence how the different stages of deleveraging develop. To cite a relevant example, the quality of institutions plays a vital role in these stages: institutional quality has a decisive impact on how private debt is restructured.

4. See Reinhart, C. and Rogoff, K. (2010), «Growth in time of debt», American Economic Review Papers & Proceedings, No. 100, p. 573-578.

5. See, for example, Baum, A., Checheriat, C. and Rother, P. (2013), «Debt and Growth: New Evidence from the Euro Area», Journal of International Money and Finance, Vol. 32 p. 809-821, which estimates a threshold of 95% of GDP.

6. See Pescatori, A., Sandri, D., and Simon, J. (2014), «Debt and Growth: Is There a Magic Threshold?», IMF Working Paper WP/14/34.