Is public debt a cheap lunch?
The fiscal and welfare costs of public debt, following Olivier Blanchard's presidential lecture at the American Economic Association, in which he suggested both might be lower than expected. We review his paper, along with several scholars' comments, and provide a quick comparison with the European context.
Olivier Blanchard sparked a vivid debate with his last presidential lecture at the American Economic Association, focusing upon the costs associated with public debt. In a nutshell, the former IMF chief economist argues that, under certain conditions, higher public debt might have limited welfare costs, and no fiscal costs at all.
The latter essentially depend on the dynamics of r (the nominal interest rate paid by the government on its debt) and g (the nominal growth rate of the economy). Contrary to widespread beliefs, Blanchard shows that g has been higher than r for most of US history. Currently, 10-year US bonds yields are in the neighbourhood of 3%, whereas forecasts of nominal growth are close to 4%. In the case of the euro area, the gap is even wider (the 10-year euro nominal rate, an index constructed by the ECB by aggregating different sovereign bonds, gives a return of 1.2%, compared with a 3.2% 10-year nominal growth forecast). If this inequality were to hold, it would imply that governments could roll-over their debt without having to raise additional taxes, as their intertemporal budget constraint would vanish.
The associated welfare costs depend on the limit that greater public debt poses to capital accumulation. More specifically, the welfare cost of higher public debt rises with the marginal product of capital, which in turn depends on the average safe and risky rates. According to the French economist, in the current context the net effect may well be positive.
Some elements further question the emphasis usually put on public debt reduction. First, estimates of the fiscal multiplier have been higher in the aftermath of the financial crisis, due to (i) borrowing constraints amplifying the effect of current income on spending and (ii) restricted leeway for central banks with policy rates close to zero. Second, hysteresis effects (i.e. the persistent impact of short-run fluctuations on the long-term potential output) suggest that a temporary fiscal expansion during a contraction could even reduce debt on a longer horizon. Third, public investment in particular has suffered from fiscal consolidation across advanced economies, despite its potentially high marginal product. Fourth, the secular stagnation hypothesis sketches a scenario where public deficits might be persistently required in the presence of negative rates in order to sustain demand and output growth.
Blanchard lists some counterarguments too. Safe rates may be kept artificially low due to excessive financial regulation, for instance in the form of liquidity requirements. Also, the future behaviour of safe and growth rates might differ from those observed so far, thereby undermining the kernel of his argument. More importantly, the presence of multiple equilibria at high levels of debt might increase the chances of self-fulfilling prophecies from investors who consider the levels of debt excessive, thereby requiring higher risk premia, which in turn would make debt-servicing more expensive and, in the end, possibly unsustainable. However, it is far from easy to assess the optimal level of debt, i.e. low enough to be shielded from such vicious spirals, and most advanced economies would likely be past this point by now.
Interestingly, a quick look at the four largest European economies captures a significantly different picture, especially for what concerns Italy (and, to a lesser extent, Germany). Although OECD data on Germany and Italy is only available from 1992, the figure above shows that r has been consistently higher than g in Italy in the last two-and-a-half decades (with the only exceptions being 2000 and 2017), whereas only after the crisis has the spread between the two German rates been positive. France and the United Kingdom, for whom data is available from 1961 onward, exhibit a slightly different behaviour: the former has an average difference between g and r negligibly higher than zero, whereas for the latter the positive gap is wider than the American one (0.45 vs 0.32 percentage points respectively). Of course, a crucial element that is not included in these charts is the debt size, with its consequences for the relevance of the interest rate and debt sustainability.
Blanchard’s goal was “to allow for a richer discussion of debt policy”, and judging by the immediate reactions to his lecture, he did not miss the target. Paul Krugman, on his New York Times blog, argues that Blanchard’s paper vindicates post-crisis anti-austerity stances and provides additional evidence that fiscal austerity might be the wrong policy answer also in the proximity of full employment – although it is true that in such labour market conditions, high debt favours current consumption expenditure over forward-looking investment. Self-reinforcing debt spirals are not meant to take place if the interest rate is indeed lower than the growth rate of the economy, even at high debt levels, and the currently observed low rates of return on private investment suggest that policymakers should focus on the patent deficiencies of public infrastructures.
The Committee for a Responsible Federal Budget criticises Krugman’s perspective, as Blanchard’s message solely implies that nominal debt can keep growing with a simultaneous decrease in debt-to-GDP as long as the economy at large is growing too, and that a small primary deficit could be sustainable. However, the American government is projected to run a large deficit this year (close to 4.6% of GDP) and in the years to come, jeopardising long-term debt sustainability. Furthermore, forecasts concerning health-related and retirement expenditure show that American public finances will be put under increased stress in the near future, so that substantial interventions on welfare programmes and the tax code would be required in order to balance the books. Blanchard’s analysis, concludes the CRFB, is theoretically correct, but has no direct relevance to the current US scenario, where public finances are far from running small primary deficits, let alone a primary balance, and where the debt-to-GDP ratio might reach 152% of GDP in three decades under current law, according to the organisation’s projections.
Simon Wren-Lewis recognises the importance of Blanchard’s lecture and focuses on the r < g inequality. Many economists assumed that, in the absence of financial repression, real interest rates would have been higher than growth, thereby leading to the intertemporal budget constraint requiring a sovereign to raise taxes in the future to finance additional debt today. However, in the current secular stagnation-alike scenario, this trade-off vanishes: a government can issue debt, repay its interests by borrowing and see its debt-to-GDP ratio decline at the same time, thanks to economic growth. This, however, concerns one-off increases in debt, as persistent primary deficits would still increase the debt-to-GDP ratio.
Thomas Philippon, from NYU Stern, concurs that social costs of debt are lower than usually thought, and that focus should be on public investment or tax reduction rather than on the pursuit of primary surpluses. Nevertheless, this prescription should only apply to (i) countries where the growth rate is indeed higher than the interest rate (which is not the case, for instance, in Italy) and (ii) high-quality public expenditure, as opposed to mere unproductive giveaways.
In his Project Syndicate column, Kenneth Rogoff argues that a rise in both short- and long-term interest rates is unlikely but not impossible in 2019. Factors that could contribute to such an outcome include (i) a positive shock to productivity, if the digital revolution finally started to more visibly affect economic figures, (ii) declining growth rates across Asia, which could transform the continent’s surpluses into deficits, and (iii) the worldwide rise of populism, fostering a reassessment of risk premia and safe rates on government debt due to accrued political risk. Although low rates indeed indicate that higher debt levels are sustainable, Rogoff stresses that additional debt still is not a free lunch, as it reduces the margin of manoeuvre for governments in the presence of shocks, thereby increasing the risk of long-term stagnation.
Government financing is also the subject of Noah Smith’s opinion piece on Bloomberg, which also touches upon some aspects of the Modern Monetary Theory (MMT), a minoritarian but increasingly popular economic approach – especially on the left side of the political spectrum (see the recent endorsement by the American Democrat congresswoman Alexandria Ocasio-Cortez). According to this theory (explained in greater detail by Josh Barro), larger budgetary deficits are not an issue, insofar as they can be financed by the central bank. The only limit to a sovereign’s capacity to finance its debt service would thus be represented by the (over)utilisation of real resources. Such a framework would require the abandonment of one of the central tenets of current advanced economies, namely the independence of central banks, since the latter would be forced not only to closely coordinate with the relevant Treasury, but also to keep interest rates close to zero and, in extreme scenarios, to directly purchase mortgage and corporate debt. The main risk would then be represented by the inability to avoid (hyper)inflation in the case that a government keeps spending in a full-employment scenario – a risk probably not worth taking, according to Smith.
MMT advocates explain that tax increases should be given the role of keeping inflation under control. This is considered one of the major weaknesses of the approach even by some of its supporters, as the political incentives to raise taxes in order to exert downward pressure on inflation would be limited at best. According to Brendan Greeley’s post on FT Alphaville, however, American policymakers, in particular Republican ones, have already integrated an approach to budgetary issues very close in spirit, consciously or not, to that championed by the MMT: spending is increasingly disconnected from taxation, and balancing the books became a temporary priority only in the late 1990s, when the cost of debt servicing had worryingly increased.
Desmond Lachman, on the Wall Street Journal, associates the MMT with Blanchard’s message, as they would both be based upon the same false assumptions: that rates will stay low indefinitely and that, as long as g is greater than r, the rest does not matter for debt accumulation. In fact, the persistent and sizeable primary deficit of the American public sector (around 3.5%) implies that the currently observed 1% margin between interest and growth rates would not prevent the US debt-to-GDP ratio from increasing. A recession would further boost the deficit, as well as any infrastructure plan not funded by taxes. In addition, the announced reduction in the Federal Reserve’s holdings of public debt, coupled with a possible increase in the risk premium required by investors scared by piling American debt, could well mean that the resulting interest rate on government bills will go up.
The discussion on the optimal level of public debt and on its relative costs vis-à-vis taxation is (at least) as old as David Ricardo, and Blanchard’s lecture will not suffice to settle it (nor does it have the ambition to do so). Nonetheless, it reignites a debate of utmost importance, especially in the current context of dismal growth prospects. The Bloomberg editorial board, for instance, already asked the German government to start preparing expansionary fiscal plans, were the observed economic slowdown to persist. David Lipton, the IMF’s First Deputy Managing Director, stresses that, in the case of a new recession, the current higher levels of public debt would translate into a smaller scope for fiscal stimuli than 10 years ago. Getting the size of the available fiscal margins right is thus essential in order to prepare appropriate policy responses.
Republishing and referencing
Bruegel considers itself a public good and takes no institutional standpoint. Anyone is free to republish and/or quote this post without prior consent. Please provide a full reference, clearly stating Bruegel and the relevant author as the source, and include a prominent hyperlink to the original post.