My Blog

September 15, 2007

On September 11 Ben Bernanke, Chairman of the Federal Reserve, gave a very useful presentation at the Bundesbank Lecture in Berlin.  It can be read at http://www.federalreserve.gov, and I strongly recommend that my Peking University students all read it.

Bernanke argues, as he has many time before, that the world is experiencing a savings glut.  According to him a number of developing countries, especially China and the OPEC countries, along with Japan, are saving far more than they are investing.  That means inevitably that they must export capital and run trade surpluses.  As the US is usually considered the safest and deepest financial market in the world, the US is the recipient of the world's global excess savings.  The inevitable result is that the US must run substantial trade deficits as the counterpart to its capital surplus.

I have been a believer of this thesis for several years.  Unfortunately Bernanke’s position has become much politicized and there are arguments back and forth about whether his hypothesis is merely an attempt to "blame" the US trade deficit on excess savings by foreigners rather than excess consumption by Americans.  These sorts of arguments are idiotic.  The fact is that any deficit country must by definition have “excess” consumption over savings, and any surplus country must have ”excess” savings over consumption, and it is not at all obvious which way the causality runs.  At any rate in my opinion this global “imbalance” is probably a good thing in the long term because running trade surpluses against the US is the only way Europe, Japan, China and Russia will be able to pay for the very brutal demographic adjustments they must make over the next two to three decades – and make no mistake, these adjustments will be brutal.

But leaving aside the silly argument as to whose fault it is, does Bernanke's argument do a good job of explaining the current US-China balance of payments?  I think the answer is yes – in fact it seems to me that China is a particularly good example of Bernanke’s thesis.  China does save too much – even the Chinese authorities acknowledge this, and they have made repeated and unsuccessful attempts to boost consumption.  The resulting trade surplus with the US is the inevitable consequence of this excess savings.

There has been a series of decisions made at both the macro level and at individual levels that explain the high savings rate in China, and these decision lead inexorably to the accumulation of US assets (through central bank purchases). Of course if China is running a large capital account surplus with the rest of the world, it must run an equally large trade surplus, and as the only country capable of absorbing such large flows, it falls to the US, with its very open financial markets, to absorb China’s trade surplus (excuse me for fudging the distinction between a trade surplus and a current account surplus, but in this case the distinction is unnecessary).

Of the three most obvious reasons leading to this decision towards excess savings, in my opinion, the first and most obvious arose as a consequence of the Asian Crisis in 1997. China’s policy-makers, like those of many other countries, were horrified by the impact of the crisis on the affected countries.  Unfortunately; also like many other policy-makers, they may have drawn the wrong conclusion about the cause of the crisis.

The Asian Crisis, like all financial crises, was caused because of serious mismatches in the national balance sheets that left the afflicted countries vulnerable to shocks that could quickly cause their balance sheets to unravel.  These mismatches create what looks like a virtuous circle when conditions are good, but they quickly become vicious circles when conditions change.  In the case of Korea, Thailand, Indonesia and Malaysia in 1997, this mismatch occurred in the form of having used highly liquid external capital for many years to fund less liquid domestic assets.  As long as capital poured into these countries, as they did until 1997, the result was a boom in which both sides of the balance sheet improved simultaneously – domestic asset values rose while real appreciation in the value of local currencies eroded the cost of external debt.  The process led to imbalances in asset values, however, which ultimately would cause capital to flow out – to devastating effect on the national balance sheets.

The incorrect lesson learned was that it was too much external debt and the lack of foreign currency reserves which left a country vulnerable to crisis (this lesson, of course, merely seemed to reinforce the lessons of earlier crises in Mexico, Brazil and elsewhere). The policy conclusion was that countries should limit highly liquid forms of capital inflow and systematically run trade surpluses to build the necessary reserves to protect them from the risks of future outflows.  Unfortunately in their haste to implement policies that encouraged trade surpluses, financial authorities in China and elsewhere may have put into place the policies that led to equally severe, but largely domestic, balance sheet mismatches.  (I strongly believe that next big round of global financial crises will be domestic banking crises.)

The second obvious cause of Chinese macro policies to boost savings was its very Asian reliance on export growth, and import constraint, to achieve sustainable growth in employment.  Since exports are the excess of production over consumption, in a growing economy boosting net exports is the flip side of raising the total amount of savings.  I think Joan Robinson’s investment multiplier explains how this happens.  As Chinese authorities channeled investment into infrastructure and production facilities aimed at developing the export sector, the resulting increase in national income was separated into high savings and low consumption, and the growing difference between production and consumption was exported.

This decision to boost exports had particularly Chinese reasons.  While most state-owned enterprises, which had dominated the economy until very recently, were inefficient and had too many useless workers, the export sector could take advantage of China’s natural advantages – cheap but dependable labor, a relatively strong infrastructure, and highly concentrated economic policy decision-making – to fuel an export boom that would absorb workers. 

Among the policies put into place to support export growth was an undervalued currency within a rigid currency regime – it had to be rigid to reduce uncertainty among exporters.  This export-orientation was exacerbated by the decision to join the WTO which, in my opinion was not about the benefits of free trade (the Chinese government has no natural predisposition to free trade) but rather about the need to use external constraints to open up the domestic markets, which were subject to a host of impenetrable trade barriers among provinces.  In that sense I liken joining WTO to Argentina’s use of a currency board (an external constraint) to force discipline on the spending habits of provincial governors.

The third obvious source of excess savings was the transformation taking place in China that significantly increased uncertainty as it reduced the social safety net.  As the cost and need for education rose, as medical services collapsed except for those with money, and as it became clear that there would be no protection for those that retired or were put out of work, worried Chinese families put an increasing portion of their rapidly rising income into savings, in an attempt, not yet wholly successful given the pace of the safety net collapse, to protect themselves from uncertainty.

Comments (7) for "China and the savings glut (1)"

I generally agree with the idea of a "savings glut" but there are a lot of details that I strongly disagree with. 1) The first is that the standard narrative of Chinese decision making makes things much more coordinated and intentional than they actually where. I don't think that there was ever an intentional decision by the Chinese leadership to boost exports, run a current account surplus, or reduce the savings net. These were unintended results of policies that started in the late-1980's and early-1990's. The currency peg was instituted in 1993, and the decline in Chinese social services was the result of inaction rather than action. The specific policy decision that lead to the current state of affairs was the decision by Bush to cut taxes without reducing social spending while at the same time pursuing a major war in Iraq. That the Chinese government did not instantly react to this is more of a case of (necessary) bureaucratic inertia than intentional decision. It simply takes about three to five years to make and implement decisions of these magnitudes. 2) The second is that I do not believe that China, as a whole. is domestically overinvesting. China, unlike Japan of the 1980's, is massively underdeveloped, and in order to bring China to first world standards of living, there needs to be a huge amount of investment. Also, I do not believe that there are huge balance sheet mismatches on the books of the banks, and I've commented on the reasons why elsewhere. But one factor that gets overlooked in comparing China to other nations is that China is more of a continent than a nation-state. There is not a lack of places within China that need capital investment. There certainly is far too much money going into the Shanghai stock market and Shanghai real estate, but the booming coastal areas are a small part of a much, much larger economy. 3) Which gets to the final point. The way to deal with the savings glut is not to boost consumption, but rather to make the domestic capital markets more efficient in order to effectively deal with the massive savings and retirement situation that is coming up. In the absence of good domestic capital markets, the solution then is to allow more investment in overseas capital markets which can efficiently process the savings. By Joseph Wang - 9/15/2007 9:00 PM

Joseph, We'll have to agree to disagree about balance sheet mismatches, although one point I want to make is that the serious mismach is not at the bank level but rather at the macro level. It is the national balance sheet that I worry about, although obviously the hole in the banking balance sheets and the self-reinforcing mechanisms embedded in their operations (a large increase in risk or defaults leads to hoading, which leads to more defaults), plus their domination of the capital intermediation process, are at the heart of the national problem. But you make an interesting point about China's being a continent. One of the standard refrains in my investor presentations is that we cannot think of China as a single clearing system. It consists of many different parts that don't always interact smoothly and without friction, and it is both possible and even likely that numbers which seem reasonable at a macro level actually conceal severe imbalances. In that sense it is perfectly possible for China to experience both rising unemployment and a worker shortage in many industries, just as it is perfectly posssible for China to have massive overinvestment and masssive underinvestment, or for certain sectors to experience inflation while others experience deflation. There are times when averages are irrelevant because they hide imbalances that may work in opposite ways in different sectors of the economy. By Michael Pettis - 9/16/2007 8:56 AM

To give an example of that we can not think China as a single clearing system is that the Policy of Developing the West. (Although we don’t want to admit the truth, that policy was failed and has no big effect on the west development.) We usually divide China in to East and West. East, especially the east coasts, stands for the rich parts and west for the poor. Several years ago, our government acclaimed the Policy, which encouraged investing in the west. By which, the government tried to narrow the gap between the east parts and the west. Under this circumstance, massive of money flew into the west. But after several years, that Gap has becoming bigger and bigger. For one reason, most of the western government which received oceans of money couldn’t find place to invest. Most of the investment, which we called Policy Investment, went back to the east coasts in order to find better return. According to the investment which ran back to the east, the money that stayed in the west was small. By Ming Gao - 9/16/2007 8:53 PM

Something similar happened with a lot of third world development aid. It was loaned to the poor countries of the world, ended back in the hands of the rich countries, leaving nothing to the poor countries except for debt. The problem is that money attracts money. Large amounts of money and large projects requiring large amounts of money in particular require expertise which are located in richer areas of the world. The possibility that a mismatch somewhere will lead to major problems is precisely what I'm worried about. In particular, I take it as given that some financial institution in China is doing something very stupid and that will explode in someone's face in a while. The question is does the system have enough reserve capacity to avoid having things spin out of control. This is hard to say, but this calls for trying to fix the problems that you can see, and strengthening the system overall, so that when the storm hits, you have some reserve to deal with it. One should note that the financial system rural China has problems far worse than anything likely to be found in urban China. The rural credit cooperatives are broke, and it's only in the last two years or so that people have even begun to start fixing these problems. By Twofish - 9/17/2007 6:25 AM

1) I agree with Joseph's point that the tremendous current account surplus and the economy's boosting export are not what Chinese leadership intentionally want, although the government contributed a lot at the very beginning of reserve accumulating, especially after the Asian Crisis. 2) Joseph mentioned the " make the domestic capital markets more efficient in order to effectively deal with the massive savings and retirement situations..." Admittedly, the maturity of Chinese capital market is extremely important, but I don't think it can help here, since investment scale will not change no matter the investment comes from domestic investor or FDI. 3) According to Michael's analysis, the underlying problem to China is the undervalue of RMB. Some tough adjustments are needed to stop unusual monetary expansion. But I'm keeping on wondering that is it possible to find an approach to ease the situation and make the eventually will-come horrible adjustments not that awful. Since the excessive production roots in the rising trade surplus and augmented by huge FDI. Assume the trade surplus can not be changed immediately and efficiently. What can be controled is FDI,especially those speculative hot money inflows. If we can control speculative capital inflows (the amount of this kind of capital is said to be huge), and meanwhile, if the constraint on oversea markets investments is loosened as well,which will increase the capital outflows. The domestic investment growth will shrink, that will lead to a lower growth rate in production development. This probably will not reduce the trade surplus, but can prevent the "imbalance" going too wild. By Jerry Wu - 9/18/2007 9:56 PM

I think speculative and investment money inflows come in because Chinese assets are relatively cheap thanks to the low currency value and are expected to rise in dollar terms. The only way to "fix" that problem is to remove expectations of a rise in value. This is being partly accomplished by higher inflation in China than in the US, but it may be temporary, and even if it isn't, it will take many years (which China doesn't have, in my opinion) to fix it. The only other solution seems to be th raise the value of the RMB enough to encourage outflows. By Michael Pettis - 9/19/2007 11:35 AM

FDI is a very small amount of the total investment in China. The basic problem I think is that China has too much money for it to efficiently handle. What should be happening is that all of those savings should be channeled to less developed parts of China, and there is no shortage of places in China that desperately need capital. The problem is that because the markets are being overwhelmed, the savings are being poured into "stupid things" like Shanghai stocks. The trouble with controlling speculative inflows is that there are basically no capital controls on inflows, and not enough time to develop those systems. The Chinese government never had an explicit policy of encouraging exports but they have had explicit policies of encouraging foreign investment and maximizing industrial output, and so there are basically no controls about bringing foreign money into China and very few controls designed to reducing industrial output. Whether "encouraging exports" was an explicit policy or an unintended consequence might seem irrelevant, and in the standard macroeconomic equilibrium analysis it is irrelevant, but from a public policy standpoint it is very important. The issue is that policies take a long time (years to decades) to change, and so policies that made a lot of sense in 1998 are now I think economically counterproductive. The problem in coming up with new policies is that those take a long time to come up with and implement and so things that might make sense today, may be absurd five years from now. We are now living in an era in which the economic environment is changing much faster than institutions and policy can change, and policy has to reflect that. This is why it is really necessary to have an accurate impression about what is going on. "Catastrophic success" is a major danger. It's not "Ha Ha. We fixed the problem and everything is perfect" it's "Oh no! We fixed the problem, things are somewhat better than before, but the fix brings with it new problems." By Twofish - 9/19/2007 10:14 PM