VoxEU Column Global crisis

The source of the global trade imbalances: Saving glut or asset price bubbles?

Global imbalances are seen by some as contributing to the global crisis – but what caused the imbalances themselves? This column argues that the popular savings glut hypothesis appears to be at odds with the data. Instead a behavioural explanation based around asset-price bubbles is a much better match for the key facts.

In 2005 Ben Bernanke gave two influential speeches where he launched the hypothesis of a global savings glut as the cause of the world’s large trade imbalances in general and the large US current account deficits in particular. (Figure 1 presents the global imbalances clearly.)

By “global savings glut” Bernanke meant a significant increase in the global supply of savings, starting in 1995. He put forward China as one of the most important suppliers of savings during this period. Bernanke further argued that this saving glut was the underlying factor for the large inflows of credit into the US, something which in turn gave rise to a consumption boom and subsequent large current account deficits.

Bernanke’s statement was controversial, not least since it appeared to place the underlying cause, and hence in some ways the responsibility, for the US imbalances outside the US.

Figure 1. Current account balance as a percentage of GDP, US, China, OECD and BRICs

Source: Economist Intelligence Unit, OECD, BEA-NIPA, State Administration of Foreign Exchange (China)

What is wrong with the saving glut hypothesis?

There are however, some aspects of Bernanke’s hypothesis that do not make perfect sense. Even though it was definitely the case that the US current account deficit increased significantly between 1995 and 2005 (from around 1% of GDP in 1995 to 6% of GDP in 2005 according to the BEA-NIPA accounts) there seems to have been no increase in the global supply of savings during that period. In fact, according to the IMF, the global savings rate went down from 22.3% in 1995 to 20.6% in 2002 and only thereafter increased to 22.8% in 2005. It is hard to see how this can be described as a significant increase in the supply of world savings.

In a recent paper (Laibson and Mollerstrom forthcoming), my co-author David Laibson and I make further attempts to assess whether the saving glut hypothesis fits with reality. We build a model where one country (the US) receives exogenous capital inflows which are calibrated to match the US current account deficits for the period that Bernanke was focusing on.

Our model shows us that such a course of events should indeed lead to increases in US consumption. However, we also find that the investment rate should have risen by at least 4% of GDP. Intuitively this makes sense; if the Chinese government exogenously loaned US households a trillion dollars, those US households should have chosen to invest a substantial share of those funds to help make the interest payments.

But such an investment boom did not occur. Data from the BEA-NIPA accounts tell us that the US investment rate fluctuated between 1995 and 2005, ending in 2005 only 1.6 percentage points higher than it started. Moreover, just two years before Bernanke formulated the saving glut hypothesis, the investment rate was lower than it was in 1996.

A behavioural alternative

There exists an alternative explanation for the unbalanced financial flows of the period 1995-2005 that does a better job in matching the data. The model is behavioural in the sense that it does not assume that agents are fully rational and always capable of perfect optimising. Here, asset price movements are key. We argue that it was the bubbles in equity markets and residential real estate markets that made the US households feel wealthier, hence giving rise to the consumption boom.

We build a model where we introduce an exogenous asset price bubble during the period 1996 to 2006. Assuming the existence of such a bubble is far from uncontroversial, but we are building on the growing literature that claims that such bubbles do exist (such as Schiller 2005 and Glaeser et al. 2008). When modelling the bubble, we think about it as the discounted value of productivity gains that are anticipated to occur at some date in the future. The bubble bursts when this fails to materialise.

Our calibrated bubble model matches the key facts in the data in the sense that it quantitatively gives rise to extensive increases in consumption and very modest increases in investment that, taken together, lead to the large increase in the current account deficits, with a peak at 6% of GDP.

Even though we calibrate the model for the US, it has the same implications for other countries. The framework implies that countries that experience an asset bubble should also experience a consumption boom and a commensurate current account deficit. This relation seems to hold empirically and has also been studied by, among others, Aizenman and Jinjarak (2008). The figure below depicts the cross-sectional relationship between country-level house price appreciation and the cumulated current account surplus. That there is indeed a strong negative relationship between housing price appreciation and the current account surplus can also be established by using a weighted regression.

Figure 2. Housing price appreciation vs. current account accumulation, 1996 – peak of housing market

Source: Economist Intelligence Unit, OECD, IMF, State Administration of Foreign Exchange (China), De Nederlandsche Bank, BEA. The area of each circle is proportional to 2008 GDP.

It should be noted that these correlations do not settle the issue of causation. There is however, in our view, significant evidence suggesting that causality runs from asset bubbles to current account deficits. For example, Greenspan and Kennedy (2008) show that the increases in housing prices contributed substantially to the consumption boom via home equity loans and mortgage refinancing cash-outs.


The large global imbalances in trade and current account have puzzled economic researchers and policymakers alike since the mid-1990s when countries like the US started to build up large deficits whereas large surpluses were accumulated in other parts of the world. Understanding the causes and effects of these imbalances is important, not least because they played an important role in the subsequent financial crisis.

The leading hypothesis of the imbalances being caused by a global saving glut has some rather severe flaws. In particular there was no increase in the global supply of savings in this period and in addition such a global savings glut would have given rise to an investment boom in the countries that imported capital, such as the US, but this was not forthcoming.

Many key facts regarding international imbalances seem to be better matched with a behavioural model that takes asset price bubbles as the starting point. Such a model can, for example, explain why the US did undergo a consumption boom, starting in the mid-1990s, whereas investments only increased marginally. The main mechanism of the behavioural model is that the asset bubble makes household feel richer, whereby the consume more and lower their active savings.

Even though there are several puzzles remaining and much more research is needed, we believe that this behavioural approach will prove to be very fruitful. It will also have significant policy implications, offering insights on when asset-price bubbles can no longer be ignored.


Aizenman, J. and Jinjarak, Y. (2008) “Current account patterns and national real estate markets” NBER Working Paper 13921.

Bernanke, Ben S (2005a), “The global savings glut and the US current account deficit: Remarks by Governor Ben S. Bernanke at the Sandridge Lecture, Virginia Association of Economics, Richmond, Virginia”, The Federal Reserve Board of Governors.

Bernanke, Ben S (2005b), “The global savings glut and the US current account deficit: Remarks by Governor Ben S. Bernanke at the Homer Jones Lecture, St. Louis, Missouri”, The Federal Reserve Board of Governors.

Glaeser, Edward L, Joseph Gyourko, and Albert Saiz (2008), “Housing Supply and Housing Bubbles”, mimeo, Harvard University.

Greenspan, Alan and James Kennedy (2008), “Sources and uses of equity extracted from homes”, Oxford Review of Economic Policy, 24(1):120-144.

Laibson, David and Johanna Mollerstrom (forthcoming), “Capital Flows, Consumption Booms and Asset Bubbles: A behavioural alternative to the savings glut hypothesis”, Economic Journal.

Schiller, Robert J (2005), Irrational Exuberance, Second edition. Princeton: Princeton University Press.

2,100 Reads