The disappointing recovery of the US and other advanced economies has resurrected interest in Alvin Hansen’s idea of secular stagnation (Hansen 1939). The argument is that the industrial world is plagued by an increasing propensity to save and a declining propensity to invest. The result is a declining equilibrium real interest rate. This can create problems due to the fact that central banks cannot lower their interest rate much below zero. A defining element of secular stagnation theories is that unlike most pervious analyses of liquidity trap situations in the tradition of Krugman (1998) and Eggertsson and Woodford (2003), it is not necessarily assumed that the real interest rate will increase back to a positive ‘normal level’.
In a world of secular stagnation, the presumption of an automatic return to ‘normal’ is unwarranted as the natural rate of interest can be persistently or even permanently negative. While seemingly a somewhat far-fetched theoretical hypothesis only a few years ago, the possibility of secular stagnation now looks increasingly plausible. Average long-term interest rates around the industrial world are now lower than they were five years ago in the immediate aftermath of the Global Crisis. As this is written, financial markets suggest that inflation is expected to remain well below its 2% target for well over a decade in the US, Europe and Japan, and that real interest rates are likely to remain well below 1% into the indefinite future. Last month we learned that Germany has followed Japan into negative 10-year rates. The US is only one recession away from joining the club. Even more disturbingly, our recent research suggests that large capital inflows to ‘safe havens’ such as the US, Germany and Switzerland are likely to create important headwinds in these economies going forward. Events like Brexit and the political turbulence in Turkey can be catalysts for such developments.
Much of previous work, including our own writings (Summers 2014, Eggertsson and Mehotra 2014), focuses on the secular stagnation hypothesis in the context of the US. Our two recent papers, however, written jointly with Neil Mehrotra (Eggertsson et al. 2016a, hereafter EMS) and with Neil Mehrotra and Sanjay Singh (Eggertsson et al. 2016b, hereafter EMSS), highlight the importance of real exchange rates and especially international capital movements in spreading secular stagnation, and the resulting policy spillovers across countries.
In our first paper we consider a slightly modified variations of a basic open economy text-book model in the IS-MP tradition (see EMS), and then in the second paper a more modern open economy overlapping generation model with nominal frictions and imperfect financial integration (see EMSS). Thus while the first paper speaks directly to current textbook treatments, the second – even if somewhat more cumbersome – speaks directly to the class of models currently in used in most policy institution of the DSGE variety. Several key implications are robust across the two frameworks, and the both point toward a much larger spillovers of policies across countries than under normal circumstances.
We show that, generally, secular stagnation transmits itself between countries via two complementary channels. First, weakness in demand abroad and a binding zero bound will generally lead to higher real exchange rates at home due to weaker demand for exports, putting pressures on the central bank to lower rates to keep aggregate demand at pace with potential output. This channel is independent of international capital markets, and would be effective even if trade were balanced at all times so there is no borrowing and lending across borders. A second channel comes about due to capital flows (which is operative even if real exchange rates are constant, e.g. in the extreme case when imports and exports are perfectly substitutable). A country finds itself in a state of secular stagnation because its desired savings outpaces desired investment unless interest rates turn negative, so that if borrowing and lending is possible across borders then excess savings will generally flow from the country experiencing secular stagnation to the other via current account surpluses. This puts downward pressures on the real interest rates in the country that is the recipient of the capital inflows. The strength of this channel depends critically on the degree of capital market integration. Our two complementary policy frameworks lead us to three broad conclusions.
First, greater financial integration may work towards spreading secular stagnation. In particular, we show that capital inflows have strong negative externalities for a country that already finds itself constrained by the zero lower bound, a result reminiscent of neo-mercantilism. Inflows of capital to a country at the ZLB actually hurt it economically, or in other words it can be harmful for a given country to run current account deficits at the ZLB. At zero interest rate there is a mismatch between desired savings and investment, and capital inflows exaggerate this problem. From the perspective of the US, an immediate implication includes possible harmful effect of dollar reserve accumulation of oil-producing countries and/or emerging markets. Another key implication is that if the Federal Reserve tightens its policy ahead of its trading partners, this is likely to lead to large capital inflows to the US, which will tend to amplify the contractionary effect of any interest rate hike and – if the capital inflows are strong enough – force the Fed’s hand to cut rates again to avoid a severe recession. Finally, strong capital inflows to the US from emerging markets due to increase policy uncertainty, or from Britain following Brexit, creates headwinds.
Second, in the open economy, policies that are stimulative for the home economy can have very different impacts on other economies and on the choices available to other countries experiencing secular stagnation relative to what we see in normal circumstances. In general, monetary policies and those directed at competitiveness carry negative externalities, while fiscal policies and policies directed at stimulating domestic demand carry positive externalities. A key mechanism behind these results is a mechanism that is normally not front and centre because interest rate can adjust freely. A key consideration is the effect a given policy has on interest rate differentials, and thus capital flows between countries, which may exaggerate or make better the mismatch between desired savings and investment in a given country. But this mismatch is at the heart of the secular stagnation problem. Expansionary monetary policy, for example, tends to lead to lower interest rates in the country pursuing it, leading to capital outflows, worsening the saving investment imbalance of the trading partner and thus creating a negative externality. Meanwhile, fiscal policy, in contrast, tends to lead to higher real interest rates in the country pursuing it – triggering capital inflows thus reducing the desired savings investment mismatch of the trading partner. In a positive sense, the fact that fiscal policy benefits spillover across countries explains why the world has relied more on monetary policies relative to fiscal policies in the wake of the Global Crisis. In a normative sense, our findings point towards the desirability of a robust fiscal response that is coordinated across countries.
Third, fiscal policies in response to secular stagnation are consistent with the government’s long-run budget constraint with three considerations being central. First, they may pay for themselves (as in DeLong and Summers 2012). We verify, in EMSS, that fiscal expansions actually lower the debt-to-GDP ratio in our framework. Second, balanced budget policies like tax-financed spending or the expansion of pay-as-you-go social security have positive fiscal impacts. Third, a one-shot increase in debt will raise demand and is clearly sustainable during secular stagnation, a result that should give many policymakers pause in their push for austerity aimed at reducing public debt.
Future policy implications
Our analysis has disturbing implications for growth going forward. If, as many believe, a loss in confidence in a range of emerging markets will drive a reduction in capital inflows and an increase in capital outflows going forward, the result will be unambiguously contractionary for the industrialised world. This is because the increase in the trade deficit cannot be significantly mitigated by reductions in interest rates, given that they are close to their lower bound. Furthermore, it is likely that increased capital flight from emerging markets will also be associated with a weakening of their economies, leading to a currency appreciation in the industrial world. The bottom-line is that capital mobility may be associated with a generalised weakening of the global economy.
This situation is quite different from other situations such as the Latin American debt crisis of the 1980s or the Asian financial crisis of the 1990s, when capital outflows from emerging markets coincided with reductions in US interest rates and a resulting increase in asset values. Now, interest rates in the US have nowhere to go, with the ZLB close to binding. It is therefore perhaps not surprising that the US stock market seems more sensitive to developments in China than would be implied by historical experience or traditional open economy models.
Our analysis also carries implications for international economic policy. For the industrial world, capital inflows are likely to be contractionary in a way that cannot be offset by monetary policy. The implication is that policies to maintain the flow of capital to emerging markets through encouragement of structural reforms, provision of official finance, and debt relief are more important for domestic prosperity than is the case when secular stagnation is not an issue. The irony of course is that maintaining political support for such policies becomes more difficult in difficult times at home.
A further implication of our analysis is that, in the presence of secular stagnation, there will be a systematic tendency for countries to rely excessively on monetary stimulus relative to fiscal stimulus. To see this, imagine that a country is at the margin indifferent between fiscal and monetary stimulus, taking account of all domestic considerations. What would be the preferences of its trading partners? Clearly they would prefer fiscal stimulus which can be expansionary because it (i) falls partially on goods that will be imported, and (ii) may lead to an increase in the real interest rate that attracts foreign capital, reducing the excessive savings problem of the trading partner. Meanwhile, monetary policy will be expansionary for the country undertaking it via an increase in competitiveness and/or lower real interest rates that trigger capital outflows, with both channels working at the trading partner’s expense.
This observation resonates with the widespread view of central bankers that they are being asked to carry too much of the burden of restoring growth in the industrial world and with widespread fears about ‘currency wars’. Cooperative efforts to internalise these externalities and achieve more significant fiscal expansion have the potential to achieve greater gains in output and to help countries achieve their national economic objectives.
Delong, B. and L. Summers (2012), “Fiscal Policy in a Depressed Economy”, Brookings Papers on Economic Activity 2012(1): 233-297.
Eggertsson, G. and N. Mehrotra (2014), “A Model of Secular Stagnation”, NBER Working Paper No. 20574.
Eggertsson, G., N. Mehrotra and L. Summers (2016a), “Secular Stagnation in the Open Economy”, American Economics Review, Papers and Proceedings 106(5): 503–507.
Eggertsson, G., N. Mehrotra, S. Singh, and L. Summers (2016b), “A Contagious Malady? Open Economy Dimension of Secular Stagnation,” mimeo, Brown University.
Krugman, P. (1998), “It’s Baack! Japan’s Slump the Return of the Liquidity Trap”, Brookings Papers on Economic Activity 1998(2), 137-187.
Summers, L. (2014), “U.S. Economic Prospects: Secular Stagnation, Hysteresis, and the Zero Lower Bound”, Business Economics 49(2).