The Role of Fundamentals in Global Imbalances
The concern over global imbalances has become an effort to systematically reduce current account deficits and surpluses, feeding the protectionist narrative. However, global imbalances are driven predominantly by domestic policies and conditions, not external factors. We do not know enough about the determinants of current account balances to set out precise numerical norms. Policy-makers should pay more attention to establishing the conditions that make current account deficits and surpluses – and their mirror image, international capital flows – sustainable.
Current account deficits and surpluses are never far from the minds of policy-makers. They are at the forefront of trade tensions. Current account balances are regularly reviewed in the publications of international organizations, most notably in the IMF’s annual External Balances Report, and in the IMF’s Article 4 consultations with member nations. The US Treasury is mandated by Congress to review the policy of trading partners twice a year, essentially to identify cases of currency manipulation.
In this brief, I will argue that it is overwhelmingly domestic fundamentals that drive imbalances, not external factors, requiring a broad and eclectic approach to evaluating them. The present emphasis on current account balances is excessive, and setting numerical norms for current account balances, as the IMF does in its External Balances Report, is questionable given our state of knowledge. For example, the effect of demographic trends on current account balances is difficult to identify with any precision. Though large current account surpluses and deficits carry obvious dangers, they can also be a good thing, justified by fundamentals and enabling capital to flow where it is most needed. Policy makers should not assume that the overriding objective is to make current account balances smaller.
Imbalances are all, or nearly all, about domestic policies and fundamentals.
I will briefly review three cases of severe imbalances which I believe help support this point. Case 1 is China-US imbalance during the years around the global financial crisis. At that time, China’s account surplus represented approximately 10% of GDP, while the US had around 6% or 7% deficit. In the years since, fiscal stimulus, a pro-wage policy and an appreciated Yuan (RMB), have meant that the Chinese savings rate increased sharply, and the current account surplus has essentially vanished.
I did not believe then, and do not believe now, that reducing the Chinese current account balance would do much to stimulate demand in the United States – simply because, arithmetically, the effect on the giant and relatively less export-dependent US economy would be minuscule under any plausible scenario. I also believed then, and believe now, that the real appreciation of the RMB would do little for the United States because the US and China are complementary economies and do not compete much directly. In fact, an RMB appreciation may even have a small negative effect on the US because it would worsen the latter’s terms of trade.
As it happens, the US did adjust in a major way – as it needed to do - since its household savings increased in the wake of the financial crisis, and there was a small moderation of domestic investment. Helped by new shale oil extraction techniques, the current account deficit of the US shrank to 2-3% of GDP. Case 1 illustrates how domestic adjustment in China and the US corrected imbalances that were at their root domestic, and did not affect each other much.
Most recently I looked at Argentina and Turkey. This is case 2. Both have lost macroeconomic control for different reasons, mainly related to misguided fiscal policy and excessive foreign borrowing. It is important to note that the crises in Argentina and Turkey are occurring against a background of solid world trade growth and short-term international interest rates that are still zero or negative in real terms. Argentina and Turkey certainly cannot rely for their adjustment on an acceleration of world demand beyond its current considerable pace.
Case 3 relates to the surplus of Germany. At the time the Euro crisis was most severe (it has not gone away) I concluded, as did many others, that Germany should reduce its domestic savings and increase domestic investment. I could see some German interest in so doing, but my main concern was the viability of the Euro-zone. The problems of Italy and Greece are not made in Germany, but an expansion in Germany would help stimulate domestic demand throughout Europe and also relieve upward pressures on the Euro, helping countries that have lost both fiscal and monetary space and cannot devalue vis-a-vis their main trading partners. Without the Euro, Germany’s real exchange rate would have moved far higher. Case 3 is a genuine instance where international coordinated action is needed to address a common problem within the Eurozone. The appropriate forum to arrive at this kind of coordination is in the Eurofin, the council of Eurozone finance ministers.
The main point I make based on these three examples is that imbalances are – overwhelmingly - made, and resolved, at home (in the case of the German surplus, “home” is the Eurozone). In this sense, they do not represent a “global” problem, but rather a domestic one. This should not be read as a message that international coordination at the G20 or IMF Board level is unneeded. Such coordination played a crucial role during the worst of the financial crisis, for example, and is needed to address a host of structural issues, such as tax competition, protectionism, and climate change.
The emphasis on containing current account balances is excessive
Given the centrality of domestic fundamentals, placing excessive attention on current account balances has a number of drawbacks. First, it makes it look like we are all playing a zero-sum game, that the “solution to my deficit is the reduction of your surplus.” This perspective feeds the mercantilist narrative that is all around us. A focus on global imbalances can provide an alibi to avoid taking difficult decisions at home, or it can create scapegoats. A focus on bilateral current account balances are especially dangerous because, in the presence of pervasive global value chains, they do not reflect value added. Aggregate current account balances can also give misleading signals. There have been times when China was clearly growing too fast and investing too much even as it ran a current account surplus. There have been times when the United States badly needed to apply fiscal and monetary stimulus even as it ran a current account deficit. Japan’s public debt is the largest in the world in relation to its GDP and requires fiscal consolidation, even as it runs a sizable current account surplus, and so on.
Current account balances are a form of trade, i.e. intertemporal trade, allowing countries to smooth aggregate demand and adjust to all manners and possibilities of internal and external shocks over time. Without current account imbalances, the flows of real capital across countries are heavily constrained, despite the fact that large amounts of financial capital flows freely. Financial flows that take the form of portfolio bonds and equities, foreign direct investment, and bank lending, are currently over 3 times larger than current account balances, which are a measure of real flows of capital across nations.
I especially question providing precise numerical norms for countries’ current account balances, to the first decimal point, as the IMF does. Our ability to quantify what determines them is limited. Various models, including those developed at the IMF, typically account only for about 50% of the historical and cross-country variation in current account balances, and there can be only limited confidence that this accounting reflects causation, i.e. has reliable predictive power. In a similar vein, one should be cautious about providing numerical norms for exchange rates. Studies have shown, for example, that currencies that are deemed overvalued or undervalued by the IMF move in the predicted direction only about 40% of the time. It is striking that in its provision of norms for current account balances, which are based on a combination of models and judgement, in no instance does the IMF provide a norm that requires a higher surplus for countries currently in surplus or a higher deficit for countries currently in deficit.
It is difficult to tell how important demographics really are in determining current account balances
Demographics play a role in determining savings and investments, and, since the current account balance is the difference between domestic saving and investment, in determining current account balances. But what role? Analysis of demographic effects typically begins with the life-cycle model of consumption, which predicts high saving rates when we work and dissaving (increased spending) when we are old, and dissaving (by our parents) when we are children, implying societies that have many old or very young people are likely to run current account deficits.However, economies are more than the sum of individuals; they are in a general equilibrium state and adjust to demographic trends in a dozen ways. Take aging for example. The life-cycle model suggests that aging societies will dissave and run external deficits. Clearly Italy, Germany, Japan and Russia – for different reasons - are not obliging. The assumption is dubious anyway since old people can decide to save more just before they reach old age, or to consume less when they are old, or they can decide to work longer or the government can decide for them, or they may even become more productive with age, and so on. Societies that are short of workers can allow immigration of workers. Importantly, aging societies also tend to have less children, exhibit slow population growth and are inclined to invest less. For these reasons, neither the conceptual nor the empirical link between aging and current account deficits has been proven in my view.
The assumption that societies that have large numbers of children will run deficits is better grounded conceptually as young societies tend also to have more rapid rates of population growth and may both save less and invest more. We do observe that very poor nations are inclined to having both large numbers of children and running large current account deficits, funded by aid, foreign direct investment, and worker remittances. However, this observation is also consistent with the theory that capital-poor countries should attract capital from capital-rich countries. And some of the deficits of poor countries reflect a desire of rich countries to help them. So even in this case, it is difficult to untangle causality and empirical studies come up with a wide range of possible coefficients (or elasticities) of current account balances on youth dependence.
What we do know with high confidence based both on theory and empirics is that increased government deficits (which are affected by demographics) tend to significantly reduce savings rates and current account balances and that oil rents tend to increase savings rates and increase current account balances. But many other variables play a role, some of which are very difficult to explain. For example, Asian nations differ enormously in income level, growth rates, demographics, resource endowments, etc. but all, or nearly all, exhibit much higher savings rates than comparators in other regions.
In conclusion, I would reiterate that, given all these considerations, we should place current account balances in a less central role and take a broader approach to imbalances, moving domestic policies to the center stage. It is also not clear why the overwhelming emphasis should always be on smaller current account balances, especially at a time of economic expansion. The present economic expansion, if it is sustained, will almost certainly imply not only increased trade, but will also lead to trade growth that is faster than GDP. There will additionally be a natural decline in ex ante saving propensities across the world and an increase in ex ante investment propensities, which may result not only in higher interest rates, but also in increased current account deficits in countries that are already in deficit. Some countries with weak fundamentals may need to contain these deficits, but many others may not need to.
Persistent large current account deficits and surpluses may also be appropriate in some instances in the long run. In 1980, Feldstein and Horioka posed a major puzzle of international economics. Why are domestic savings and investment so highly correlated, when, in principle, domestic savings should be distributed across the world to reap the highest return and the greatest diversification? They find that about 90% of the world’s increased savings is invested domestically whereas on average, taking into account the relative size of countries, it should be about 10%, were international capital markets perfect. The correlation between domestic savings and investment has declined significantly since 1980 but remains very high. The implied question is why aren’t current account balances much larger? To be sure, facilitating the implied capital flows would require reforms that make them more sustainable, including more comprehensive macro-prudential regulations and a larger IMF. But if we did succeed in this endeavor, it would be good news for developing countries that need the capital and for advanced countries that would like to see a higher return on their savings, and it would allow investors throughout the world to achieve greater diversification.