18 Responses to “Villains and victims of global capital flows”

Comments

1. Edmund Phelps: Martin´s lively piece this week is timely, since the US Congress looks poised to try to legislate the end of the “saving glut” - in China, at any rate, if not the one in the Middle East, and not to mention the “investment drought” in Germany, Italy and France, which also leads to current account surpluses.

   I have written before about what is now known as the “saving glut” hypothesis, drawing on the “structuralist models” introduced in my book Structural Slumps (1994). An improved version of one of the open economy models appears is the subject of a paper by Hoon and I in the June 2007 issue of Journal of Macroeconomics.

   This model suggests the following analysis: As incomes in China, the rest of developing Asia and the Middle East keep on increasing by leaps and bounds, a gap opens up between consumer demand and consumer good supply. (Amar Bhide and I have a 2005 paper on why fast growth tends to create such a gap in an LDC setting.) The excess supply of consumer goods seeks foreign buyers while, we suppose, no counterbalancing excess demand spontaneously arises in the west.

   If the global market for consumer goods is to clear, meaning that global consumer demand matches global supply, short-term world interest rates must drop, lifting global household wealth, to push up global consumption demand as required. The prospect of these low interest rates for a long time means that the whole real yield curve is shifted down.

   In the west, notably the US, the UK and the others to some extent, the valuations placed by businesses on each of the business assets - the employee, the customer, office space per square foot - are sharply increased. So is the value placed by households on residential structures. That leads in turn to an increase in investment demand and thus to the creation of new jobs. There is a drop in the path of the natural unemployment rate. Moreover, the drop of real interest rates boosts household consuming of all kinds, which also fits with what we observe.

   In this view, there is much in agreement with what Martin has written. There are, however, two points of disagreement.

   First, the analysis does not support Martin’s contention that the opening up of a current account deficit in the US can be credited with saving the world from a slump. If US consumers possessed less responsiveness to each one point drop of the interest rate, there would have had to be a larger drop; and Chinese consumers would have been emboldened to step up a little their consumption, about which they have been so cautious.

   Second, and more important, although this episode of “imbalance” is seen badly by Martin and some others far more critical of the market than Martin is, I find much there to be thankful for.

   Without the Chinese shock, the US economy would not come out of the slump of 2001-2002 nearly as well as it did.

   Yes, the low interest rates are not sustainable - what is? But so what? If countries spurned every opportunity that looked to be unsustainable business life would be much the poorer.

   Posted by: Edmund Phelps | June 15th, 2007 at 10:09 am | Report this comment
   

2. Richard Cooper: I agree with Martin Wolf that what I prefer to call the “excess savings” hypothesis has greater merit than the excess liquidity hypothesis. But why do the two hypotheses have to be cast with villains and victims, as if they were plots for a Hollywood movie?

   And I would add to the excess savings hypothesis the demographic dimension: rapidly ageing societies such as Japan and Germany are properly investing their savings abroad when profitable investment opportunities at home are limited.

   This, after all, is the essence of financial market globalisation. And where better to put some of such savings than in the United States, a robust economy with secure property rights, financial markets with breadth, depth, and resiliency, and a very different demographic outlook?

   Oil surpluses are of a different character and will prove, I suspect, to be more transitory as oil revenues, which accrue initially to governments, gradually enter the income stream in those countries.

   Posted by: Richard Cooper | June 15th, 2007 at 10:11 am | Report this comment
   

3. Martin Wolf: I thank Ned and Dick for their thoughtful comments.

   Ned makes two points of criticism of my (somewhat Keynesian) views. First, he argues that if Americans had not consumed more, global real interest rates would have been lower and spending elsewhere higher. I think that must be right. How powerful that effect would have been depends on the elasticity of such spending with respect to the real rate of interest.

   Second, I also agree with him entirely on the positive value of the imbalances for the global economy. Indeed, given my views, I am sure that they have kept the world economy motoring along, despite the emergence of the savings glut in China and elsewhere.

   I would also make two small comments on Dick’s response.

   The first is that the “villain and victim” phraseology is partly just journalistic licence. But it also reflects the underlying tone of the debate. Those who hold the “monetary glut” view believe that the rest of the world is the innocent victim of US monetary imperialism. Equally, some believe that Ben Bernanke was trying to blame others for the external imbalances when he launched the discussion of the savings glut hypothesis. So the loaded nouns I have deliberately used do have a certain relevance to the debate.

   The second point is that China is an ageing society, but also still a poor one. It is not so obvious why it should be a huge capital exporter. I would also add that buying US dollar bonds (the main form of capital flow in recent years) does not strike me as a great investment even for Germany or Japan. It is well known that the US has gained far higher returns on its investments abroad than foreigners have gained in the US. So large has been the discrepancy that the current account deficits have recently been something close to a free lunch for the US: huge current account deficits and yet no rise in the ratio of net foreign liabilities to gross domestic product. In other words, the deficits have been an excellent deal for the US precisely because they have been such a bad deal for many of its creditors.

   Posted by: Martin Wolf | June 15th, 2007 at 7:18 pm | Report this comment
   

4. Adrian Wood: What Martin interestingly reviews seems to me to be not, as he claims, “a contemporary version of the old Keynesian versus monetarist dispute”. It is a re-run of the older dispute between “Mr Keynes and the classics”, to quote the title of Hicks’ famous article. The “classical” view was that the interest rate adjusts to equate investment and savings. Keynes argued that the interest rate adjusts to equate the demand for money with its supply.

   Just a pedantic point of intellectual history? Maybe. But consider how Hicks synthesised these two views into the IS-LM curve diagram and its possible implications for this debate. In the diagram, the interest rate is on the vertical axis and aggregate income on the horizontal axis. The IS curve slopes down from left to right, and is intersected by the LM curve rising from left to right.

   What has been observed globally in recent years, in terms of the two variables whose behaviour this diagram describes, is a fall in the interest rate and a rise in aggregate income. Given the positions of the two curves, the only possible explanation for this combination of outcomes is a rightward shift of the LM curve. Such a shift could be caused by a rise in the money supply. No shift of the IS curve on its own could produce this outcome. So maybe there is more to the money-glut story (which actually is the more Keynesian one) than Martin supposes?

   For a more up-to-date Keynesian framework in which to think about this debate, see the new book by Wynne Godley and Marc Lavoie: Monetary Economics (Palgrave, 2007). It builds on Wynne’s long-standing and illuminating focus on each sector’s net acquisition of financial assets (the difference between its investment and its savings), rather than treating investment and savings separately.

   Posted by: Adrian Wood | June 16th, 2007 at 8:46 am | Report this comment
   

5. Willem H. Buiter: Both the ex-ante global savings glut and the liquidity flood are part of the story.

   The Federal Funds rate was kept very low for several years after hitting 1.00 percent in 2003. Japan’s policy rate was at zero for years and is still at only 0.50 percent. Policy rates of all globally significant central banks (Fed, ECB and BoJ) have been below their neutral levels since the bursting of the tech bubble. All three are below neutral even today. At least the ECB is clearly committed to a policy to end monetary accommodation and to move into a phase of restrictive monetary policy. As one would expect from this consensus-based organisation, its interest moves are excessively gradual, but at least the direction is clear.

   The Fed, on the other hand, by focusing on core inflation rather than on underlying inflation (headline inflation in the medium term), has for (at least) the past three years systematically underestimated the inflationary pressures in the US economy. This persistent analytical mistake and the policy errors it has caused are difficult to understand. Historically, US core inflation has been more persistent and less volatile than the rate of inflation of non-core goods. So in the past, core inflation may have been a useful predictor of medium-term headline inflation. Then the world changed: 2.3 billion Chinese and Indian consumers and producers became fully engaged in the global economy. The result has been a major, persistent and continuing increase in the relative price of non-core goods (food, primary commodities, energy). That means a long-lasting excess of the inflation rate of non-core goods over the inflation rate of core goods. Acting as bad statisticians who don’t realise that the data generating process of the time series they are relying on has been subject to a structural break, the Fed continues to treat core inflation as a good predictor of medium-term headline inflation. As a result, inflation has got away from the Fed and they don’t even realise it.

   I believe the day is drawing closer when the Fed will recognise the folly of their focus on core inflation as a good measure of underlying inflation and as a good guide to headline inflation in the medium run. The discrepancy between reality and the Fed’s perception of reality is simply becoming too wide and obvious. Core inflation came in at just 0.1 percent in May 20007, against a market expectation of 0.2 percent. Headline inflation was 0.7 percent in May, its biggest increase in almost two years.

   Mervyn King, in his hawkish speech of Monday 11 June 2007, said: “sharp rises in the prices of energy and food have squeezed spending on other goods and services, putting downward pressure on those prices. That is why measures of ‘core inflation’ that strip out certain prices can be highly misleading”. He is, of course absolutely right.

   King’s argument (which is further strengthened when it is recognised that the increase in the relative price of non-core goods represents an adverse terms of trade shock for US consumers), is separate from, and in addition to, another argument why an increase in the relative price of non-core goods is inflationary in the short run. This is that non-core goods are flexible price goods (set in auction markets) while core goods and services tend to be subject to nominal rigidities, reflecting long-term contracts and other institutional features. For a given path of nominal interest rates, a permanent increase in the relative price of ‘flexible price’ goods (i.e. non-core goods) will temporarily raise the headline rate of inflation.

   The inflation news from the US is therefore, in my view, bad. It confirms that the Fed has lost the plot, and has let inflation stray well above its ‘comfort zone’ and any reasonable notion of price stability. The Fed does not yet recognise this, and neither do the markets, which have followed the Fed into the core inflation blind alley, but they will. Mervyn King and Charlie Bean will probably be able to convince them that when there are major relative price changes between core and non-core goods and services, core inflation is a lousy guide to headline inflation in any run. And even if Mervyn King and Charlie Bean are insufficiently persuasive, the persistent and consistent deviation between the behaviour of headline inflation and core inflation will convince them. The Fed ultimately cares about headline inflation, not core inflation. It does not believe that Americans don’t eat, drive cars, or use central heating and air conditioning.

   Rates in the US will have to rise, not to choke off an inflationary threat, but to regain control over a burst of inflation that has already materialised. I would not be surprised if US rates hit six percent during 2008.

   The folly of the Fed is plain for all to see – plain for all except the FOMC and the Fed staff, that is. By following the will-o’-the wisp of core inflation they have let headline inflation get away from them. In the process, they have added to domestic and global liquidity and prolonged the global asset market anomalies that are just beginning to correct themselves: low long-term risk-free real interest rates and very low credit spreads across the board. They therefore have contributed to the risk that the ultimately re-normalisation of global asset markets will be abrupt and disorderly.

   Posted by: Willem H. Buiter | June 16th, 2007 at 1:04 pm | Report this comment
   

6. Guillermo de la Dehesa: Martin has written a clear and solvent piece about global capital flows and I agree with most of it, but I have a few comments to make to try to show that the savings glut, while being the main factor explaining global imbalances, may have not much to do with “villains and victims”.

   First, for the first time in more than 30 years most world economies are growing at a good pace, a clear signal that most of the world is enjoying the fruits of globalization, in spite the existence of global imbalances, there are few victims as well as villains.

   Second, there are many logical reasons that explain the savings glut:
   There are some reasons why China has such a large volume of foreign currency reserves, some of precautionary nature:
   The Chinese economy is much more open than what it is normal for any large and developing country. The sum of its imports and exports of goods and services is close to 64 per cent of GDP (imports 30 per cent and exports 34 per cent) more than double than the US openess ratio with 30 per cent of GDP (imports 18 per cent and exports 12 per cent).

   But it is true that Chinese reserve accumulation is vey high: it is about 80 per cent of external liabilities, while the world´s average is close to 30 per cent. Its reserves equal to 46 per cent of GDP and also 70 per cent of its total trade and 134 per cent of its total imports.

   Therefore, there must be a certain degree of currency manipulation which must explain the rest of the reserve accumulation, but this deliberate policy is also explained by its replication of the traditional asian export-led growth model, which has been so successful first in Japan, then in Korea and some south east asian countries and now in China.

   China is also a high saving country and it is going to keep being it, for other important reasons: Its average rate of investment, for the last decade, has been 35 per cent of GDP and its large FDI inflows have also created large external liabilities; its fast ageing population trend (a result of its previous policy of one child per family policy) is weakening dangerously its future fiscal situation, even more when there is an inadequate state-funded social retirement and health provision system which needs to be compensated for by private precautionary savings. The spiralling effect of oil prices is also a cause of great concern, not only for the government but also for the households which tend to save more and consume less.
   Oil exporting countries have aldo their reasons to save more this time. Some of them are building huge funds for future generations who will not be able to enjoy the oil revenue or stabilization funds for future swings in oil prices.

   Finally, most Asian countries are still under the trauma of the last Asian crisis and they prefer not to take any risk going forward and keep a high level of reserves to compensate for any missmatching of their foreing debt and also to avoid being forced to borrow from the IMF.
   Thus, there are many reasons why Asia and oil exporting countries are saving so much.

   Posted by: Guillermo de la Dehesa | June 16th, 2007 at 9:44 pm | Report this comment
   

7. Eric Beinhocker: As usual Martin has cut to the core of a complex issue and I would agree that the “savings glut” theory is more persuasive than the “money glut” theory for explaining global imbalances. The general equilibrium instincts of most economists would lead one to predict that what is unbalanced must ultimately come into balance. But the dynamics of global capital flows are such that we could be in an “unbalanced” world for some time to come – and this is not necessarily such a bad thing, in particular for US households.

   A new report by my colleagues Diana Farrell and Susan Lund* shows that the US current account deficit could well continue to grow, at least over the medium term (see the FT “Markets Could Take US Deficit Rise” June 15, 2007). The research models the future growth in the current account surpluses of key countries (Asia, oil producers, Europe) who are exporting the capital needed to fund the US deficit. It finds that under plausible assumptions, these surpluses could continue to grow sufficiently to fund a US deficit of up to $2.1 trillion in 2012 versus $857 billion today.

   While such a large deficit might be uncomfortable, it would not be unmanageable. Under this scenario US net foreign debt would reach 46 percent of GDP – not an unprecedented level among other countries (e.g. Mexico) – and the US would still be able to finance this large a deficit because the implied net foreign interest payments would remain at less than 1 percent of GDP.

   In all likelihood, however, the US would never reach such a level of deficit. The same analysis shows that if the world’s current pattern of savings, demand, and exchange rates persists, the US current account deficit would likely reach $1.6 trillion in 2012, or 9 percent of GDP versus 6.5 percent today. The world would thus have $2.1 trillion in savings available to cover US dis-savings of $1.6 trillion. It is therefore likely that the savings glut and US borrowing will continue at least for the medium-term.

   The one thing that could end this party is a true collapse in the dollar – my colleagues estimate a 30 percent drop would be required to balance the current account by 2012 – a level of decline that has only happened once in the past 35 years. Interestingly, even after such a drop, the US would still run a large trade deficit, particularly with China, thus showing how misplaced the US Congress’s fixation on the yuan is.

   Short of a dollar panic causing a sharp drop, the most likely scenario is a gradual rebalancing driven by long-term factors such as demographics, relative productivity growth, and the eventual shift towards consumption as China grows richer. Thus the end is not nigh for US consumers – just delayed.

   *See the McKinsey Global Institute, Diana Farrell and Susan Lund, “The US Imbalancing Act: Can the Current Account Deficit Continue?”, available at www.mckinsey.com/mgi

   Posted by: Eric Beinhocker | June 18th, 2007 at 10:49 am | Report this comment
   

8. Martin Wolf: A great many important points have been added to this forum.

   Adrian Wood tells me that the distinction I drew is not between Keynesian and monetarist ways of thinking. I derived my distinction as follows. The savings glut view is the “paradox of thrift” of Keynes, where excess savings threatens recession. The response of the monetary authorities in the Keynesian view is fiscal or monetary expansion (as in the IS-LM model he cites). So in this view the monetary loosening we have seen is the consequence of excess savings. The monetarist view of the balance of payments, however, suggests that weak exchange rates (as for the dollar nowadays) are a direct function of an excess supply of money. The latter then is the causal factor in this view. That was the distinction I tried to draw. I still think it makes sense.

   Let me, however, take this opportunity to use the IS-LM framework (which I feared to do in the column) to give what I think is roughly the right global story.

   First, there has been an increase in the world’s (ex ante) propensity to save out of income and a decline in the (ex ante) propensity to invest, partly because of the bursting of the stock market bubble after 2000, partly because of the worldwide rise in profit shares in GDP and partly because of the large shift in global incomes towards high-saving countries (i.e. East Asia and the oil exporters). All this generated a shift in the global IS curve down and to the left. If there had been no monetary policy response there would have been a recession. But there was such a response in the significant countries. So the LM curve was also shifted downwards to the right.
   The new equilibrium point kept the world economy close to “full employment” (itself, of course, a moving point in terms of output, given the expansion in global capacity), but at lower nominal (and real) interest rates than before.

   This is I think the only story that makes sense of the combination of a looser monetary policy with no significant inflation, at least so far.

   To contrast this with Adrian’s story, the expansion in output is a function of rising capacity (or rather potential output). It does not need a downward shift in LM to explain it (i.e. everything here has to be defined relative to a moving, rather than a stationary, full-employment equilibrium). So it is better to think of what has happened overall as staying close to full-employment equilibrium, but at lower interest rates. To explain that we need the offsetting downward shifts in both curves that I describe. But the downward shift of the IS curve started it. That is why the savings glut is, I think, the cause and the downward shift in the LM curve the policy response.

   Two other points are worth adding. The first is that there is not just one country, as in the basic IS-LM model, but many. The downward shift in the world’s IS curve was driven by the downward shift in the IS curves in a number of countries. This generated an expansion in the excess supply of goods and services in those countries and so a rise in their trade surpluses (or reduction in deficits). This had to be offset by an increase in demand elsewhere if “full employment” was to be sustained globally.

   There was also a downward shift in the IS curve in the US. So offsetting what happened in the world as a whole demanded a big shift in the LM curve in the US – i.e. a huge US monetary loosening - to offset not just the downward shift in the US IS curve, but the downward shifts in the IS curves of the rest of the world, too. And, of course, we also saw a large rise in the US current account deficit, as well.

   Such US monetary loosening also automatically transmits itself to any country with a currency pegged to the US dollar, thereby shifting their LM curves down and to the right, as well. Thus do we see a global liquidity glut, but in response to the global savings glut.

   The second point is that the monetary authorities can very easily overdo the loosening, given the long lags between monetary easing and inflation. Willem argues strongly and to me persuasively that the Fed has done so, by focusing on the wrong inflation indices (and ignoring the evidence on liquidity, I would add). So even if the monetary loosening was initially necessary (as I believe), it may well have been too large, to long-lasting or both. If Willem is right, we are in for a bout of stagflation, with unexpectedly high interest rates and a great deal of unpleasantness in overextended financial markets. But the latter is probably necessary, given the nonsense we can see in credit markets. Stand by, then, for the correction.

   I largely agree with Guillermo de la Dehesa on why the Asians are saving so much. Indeed, I am trying to write a short book explaining what is happening in just this way. But I am also arguing, as I think he is, too, that the savings have become excessive. Precautionary saving makes sense, but does a national saving rate of perhaps as much as 50 per cent of gross domestic product, as in China today, make sense? Reserve accumulation makes sense, too, but can reserves of $1,200bn and rising for China alone and more than $5,000bn for the world also make sense? There should be moderation in all things. This is immoderate and surely undesirable.

   I thank Eric Beinhocker for mentioning the extremely interesting McKinsey report. I used one of the charts from it. I agree that the US can finance a very large deficit, probably one much larger than today’s. It is also true that most of the risk would then be shifted to the creditors, since the US borrows in its own currency. But that raises what seems to me to be a key point: at what price might the US be able to finance growing deficits? I assume that as the share of US liabilities rises in the rest of the world’s portfolios, the returns will also have to rise, to compensate for the increasingly concentrated risks those foreign investors bear. That means higher interest rates in the US (and so lower asset prices) and a weaker exchange rate for the dollar(and so a greater probability of appreciation). Such shifts could generate financial shocks not just within the US but in a still heavily US-centric financial system. Obviously, the shocks would get even bigger if Willem is right about the threat of inflation.

   I thank the participants for what I think was a particularly good discussion.

   Posted by: Martin Wolf | June 20th, 2007 at 8:18 pm | Report this comment
   

9. Alex Grey (guest): I agree with Adrian Wood’s comment and would add that there are three key developments that have affected the position of the LM curve. These are: 1) disinflation; 2) financial deregulation and 3) financial innovation.

   The result of these three developments has been an increase in the amount of debt that an individual economic agent (for example a household) can hold for a given level of income. The effect of all three on credit is fairly simple. Disinflation, by lowering the nominal rate of interest, results in lower cash flow associated with a given level of debt and consequently increased debt holdings. The ultimate effect of financial deregulation is to increase the amount of debt that an economic agent is permitted to carry. Financial innovation has the same ultimate impact (consider innovations I mortgage lending for example). The large increase in the ratio of private credit to GDP is also reflected in a substantial increase in broad monetary aggregates such as M3 because increased credit generates increases in financial assets, some of which are included in M3. Assuming that one is using a broad definition of the Money Supply when drawing the LM curve (e.g. M3) the result of an increase in debt held by both households and firms has been a rightward shift in the LM curve. As the causes of the rightward shift in the LM curve extend back into the early 1980s, this suggests that current global imbalances have deep roots.

   Posted by: Alex Grey | June 21st, 2007 at 4:05 pm | Report this comment
   

10. Monty Graham: Last week, in the Financial Times, Martin Wolf discusses the two paradigms offered to explain the current “global imbalances”, which is to say (mostly) the current account surpluses of a number of Asian and, to a lesser degree, European nations and the current account deficit of the United States. The two paradigms are “savings glut” (Asian nations, as a percent of GDP, save too much of this income and invest or consume too little) and “money glut” (to accommodate the US current account deficit, which is home-generated, the US Federal Reserve has created an excess of liquidity; to accommodate this liquidity, the Asian countries must pursue fiscal and monetary policies that induce current account surpluses). Implicitly admitting that the two paradigms are not wholly mutually exclusive, Martin suggests that the facts as they currently stand more strongly support the “saving glut” paradigm than the other.

    If so, we must ask why is it that the Asian nations (and China in particular) have become such high savers. Focusing on China, a standard explanation is that China is pursuing a mercantilist industrial policy, and thus keeps its exchange rate low in order to generate a large surplus in traded goods and services; this in turn requires that fiscal and monetary policies (including closed capital accounts) be set so as not induce an acceleration of domestic inflation. This is a reasonable explanation and it does seem to fit well with the facts.

    It is not the only plausible explanation, however. About one decade ago, Chi Zhang completed an unpublished doctoral dissertation at Johns Hopkins University wherein he argued that the reasons for high savings rates in China and elsewhere in East Asia lay at the level of behavior of households and not at the level of macroeconomic or industrial policy. Zhang’s empirical work focused on household savings behavior in Taiwan during the 1970s and 1980s; he argued that this focus was appropriate because (1) Taiwanese experience during those decades closely approximated more recent Chinese experience and (2) extensive data were available for Taiwan that were not available for mainland China, at least not at that time. Zhang’s main finding was that, as household income rose in Taiwan, so did household savings rates. Moreover, and importantly, this was true for all cadres of households based on income (poor and rich households alike increased their savings rates as household incomes rose) and all cadres based on age (households where the head of household was young behaved no differently than households where the head of household was older). Given the latter two sets of facts, Zhang explicitly rejected that the Taiwanese behavior could be fully accounted for by the Modigliani Brumberg hypothesis, under which households would attempt to maximize consumption over a finite period of time (notably, the expected life span of the household itself); rather, the Taiwanese facts seemed more closely to support the older (but generally rejected) theory of household savings of John Maynard Keynes. The Taiwanese facts were consistent with an explanation to the effect that pension and other old-age income assurance schemes in Taiwan were less developed than in the west, but Zhang felt that Taiwanese savings rates were simply too high to be accounted for by a need to provide for old age (otherwise, the Modigliani Brumberg hypothesis would have been a better fit).

    Zhang’s own explanation for Taiwanese behavior was that it was rooted in Chinese culture. In this culture, traditionally, anything like conspicuous consumption is discouraged. Moreover, in that culture, it is expected that the current generation will pass on as much wealth as it can to the next generation; by contrast, Modigliani Brumberg assumes that households will bestow at most modest wealth, as a percent of lifetime income, on the next generation. Thus, as household incomes rose, Taiwanese households would cling to earlier “consumption habits” and hence simply save the income increases that they were receiving. Zhang notes that this persistence of “consumption habit” in Taiwan was so pervasive that, even as household expectations regarding income growth changed (i.e., households came to expect rapid annual growth), household savings rates continued to increase).

    It is truly difficult to test whether Zhang’s hypothesis explains the current high household savings rates in China better than does a more conventional hypothesis based on mercantilist industrial policy. However, if Zhang is right (or even partially right), some consequences of economic “reform” in China as sought by the United States and other western nations might not be quite as expected by western policymakers. In particular, these policymakers, especially in the US Treasury, advocate eventual capital market opening (which, as acknowledged even by US Treasury officials, must be preceded by measures to strengthen the Chinese banking system) that is accompanied by a free float of the Chinese currency. The generally expected result is that autonomous private capital flows into China would increase, putting upward pressure on the Chinese currency and hence leading to reduced net Chinese exports and increased domestic consumption or investment. However, if Chinese households were stubbornly to refuse to consume more, in spite of reduced import prices and hence increased real income, but were now able to place their savings overseas, the result could be a capital outflow from China. Rather than exchange rate appreciation, exchange rate depreciation could follow.

    Do I really believe that this latter would happen? To answer, I would have to think harder and deeper than I now have. But, for the moment, I would not rule it out.

    Posted by: Monty Graham | June 22nd, 2007 at 11:44 am | Report this comment
    

11. Wynne Godley: I agree with Richard Cooper that the “saving glut” and the “money glut” should not be seen as rival explanations for the imbalances. Surely they are the joint outcome of interdependent processes, from which both the surplus and the deficit countries have derived (what they can so far experience as) great benefits.

    I know it is fashionable to attach supreme importance to monetary vs fiscal policy. All the same, I am surprised that no-one has pointed out that the US’s general government budget balance was relaxed by an incredible 6-7 pps of GDP between 2000 and 2003. This was surely the main reason why severe recession was avoided.

    The large US current account deficit could not have come about without a large fall, into negative territory, in private net saving (that is, saving less investment). And this negative net saving could not have come about without very high lending to the non-financial private sector. But the flow of lending is fast drying up as a result of the housing debacle. If private net saving during the next few years reverts to what was once its normal level (a change equivalent to perhaps 5 pps of GDP) the US could encounter a long period of stagnation, or worse, without there being any obvious remedy.

    Posted by: Wynne Godley | June 25th, 2007 at 11:34 am | Report this comment
    

12. Richard Duncan: Martin Wolf refers to two alternative explanations for global imbalances as the “savings glut” theory and the “money glut” theory. The former he accredits to Fed Chairman Bernanke. The latter he associates with the argument expressed in my book, The Dollar Crisis: Causes, Consequences, Cures. I would like to take this opportunity to elaborate on the global money glut, because I consider the alternative explanation to be misleading and counter-productive.

    The global money glut has come into existence as the result of two related developments. The first development is the emergence of large trade imbalances between high-wage countries such as the United States and ultra low-wage countries such as China. The second development has occurred as a reaction to the first. It is the aggressive creation of paper money and foreign exchange intervention undertaken by the central banks of the surplus countries to ensure that those trade imbalances persist.

    Free trade between the United States and low-wage countries like China inevitably leads to large trade imbalances. Wages and benefits in the US amount to $200 per day compared with $5 per day in China. The large, trade-driven capital flows into the low-wage countries naturally exert upward pressure on the currencies of those countries. In order to maintain their low wage advantage, the central banks of the surplus countries buy all the dollars entering their countries in order to prevent their currencies (and, consequently, their wages) from rising. To buy those dollars, they must first create or “print” the equivalent amount of their own currencies. This kind of money creation is occurring on an unprecedented scale. For example, total foreign exchange reserves held by central banks have doubled to over $5 trillion over the past five years. In other words, central bank reserves have increased by as much in the last five years as in all prior centuries combined.

    Once central banks, such as the PBOC, have bought the dollars, they must invest them in US dollar assets in order to earn a return. This pushes up asset prices in the United States, including bond prices, and, thereby, pushes down yields. This is the explanation for Alan Greenspan’s “conundrum”. Low US interest rates and rising asset prices fuel consumption and imports and cause the US current account deficit to become even larger. This creates an expanding spiral of increased consumption and imports in the US and expanding exports, money creation and foreign exchange accumulation in the low-wage countries.

    This expansionary cycle benefits everyone just so long as it lasts. Unfortunately, it can’t go on forever. Not even the United States can continue going into debt to the rest of the world at the rate of $2 million per minute as it is currently required to do to finance its current account deficit.

    I don’t believe it is appropriate to assign blame to either the high-wage deficit countries or the low-wage surplus countries. These destabilizing imbalances are simply the outcome of all economic players acting in their own best interest in the absence of a rule based international monetary system. The “non-system” (sometimes referred to as The Dollar Standard) that emerged following the breakdown of the Bretton Woods International Monetary System is flawed because it lacks a mechanism to prevent large and persistent trade imbalances. The gold standard and the Bretton Woods System functioned in such a way that made large trade imbalance impossible.

    The global savings glut theory is inadequate and misleading because it omits the fact that central banks in the surplus countries are responsible for creating most of the “savings” by creating the equivalent of hundreds of billions of dollars worth of renminbi, yen, roubles and rupees each year, for the purpose of buying dollars and holding down the value of their currencies. Paper money creation by central banks should not be described as “savings”. The global savings glut view seems designed more to obscure the origins of the current unprecedented global imbalances than to explain them.

    Richard Duncan is a partner at Blackhorse Asset Management in Singapore

    Posted by: Richard Duncan | June 26th, 2007 at 8:38 am | Report this comment
    

13. Martin Wolf: Let me add to this debate.

    First, Alex Grey has added interesting additional comments on structural shifts in the LM curve. I do not disagree. Monetary policy is more effective than before. But this does not explain why monetary policy has become more expansionary. I would refer to my earlier discussion of the shift in the IS curve down and to the left.

    Second, Monty Graham has added an important dimension. I agree that the high savings rates of Chinese households may be structural. But it is important to remember, as I have noted in several pieces, that households are responsible for only about a third of Chinese savings, the rest being generated by corporations and the government. So I do not think that this point rules out the proposition that savings are a policy decision. I addition, I agree that the abolition of exchange controls and exchange market intervention might lead to a weakening of the renminbi as Chinese private capital floods abroad. That is certainly one of the great unknowns. Should we worry about that? If the weakening were the result of private decisions I would be relatively relaxed.

    Finally, Wynne Godley is right. Fiscal policy also mattered. In addition, if the US household sector stopped borrowing so much, the US would fall into a serious stagnation and monetary and fiscal policy would also have to be loosened again. In those circumstances the pressure for a weaker dollar might also become overwhelming.

    Posted by: Martin Wolf | June 26th, 2007 at 7:52 pm | Report this comment
    

14. Martin Wolf: This last post is on Richard Duncan’s intervention. I greatly appreciate his doing so because he is the most influential proponent of the “money glut” view. I also think the arguments he makes are simply wrong, on at least five levels.

    First, he argues that free trade between rich countries and low wage countries must lead to trade imbalances. Why? This is in violation of the theory of comparative advantage that has been the cornerstone of economists’ views on trade for two centuries. It is also inconsistent with experience.

    China’s low wages are the result of its low average productivity. (In dollars, which is what matters here, China’s average productivity per head is about a 20th of that of the US.) In some sectors, of course, China’s productivity is close to that of the US. In others it is far lower. China should export where it is relatively more productive and import where it is relatively less so. This would automatically lead to balanced trade, provided the Chinese spent their incomes.

    In reality, of course, both China and the US are embedded in a multilateral trading system. So China could have a big bilateral surplus with the US while it has a big deficit with the rest of the world. But this is a wrinkle.

    That this could well have happened is also consistent with experience. In their eras of fast growth the then “super-competitive” economies of Japan and South Korea did not run huge trade surpluses. On the contrary, they ran deficits. South Korea’s deficits were particularly large. These countries started to run surpluses only when they had become relatively rich, growth had slowed and investment rates had started to fall, relative to savings (as the savings-glut hypothesis suggests).

    Second, in the comments above on comparative advantage I assume simply that the Chinese spend their incomes, including those earned from exports, instead of hoarding a big part of them, as they now do. Spending one’s income would seem sensible and normal, particularly when one’s country is relatively poor and already investing a good 40 per cent of GDP. What is the point of accumulating mountains of dollar assets?

    This, in any case, is the basis of the absorption approach to the balance of payments, now half a century old. Provided countries spend their income, their current accounts will balance. This applies to low-wage countries and also to high-wage countries.

    In this case, we know from the national accounts that China is not spending all its income, while the US is spending a great deal more than its income. That is precisely what the savings-glut hypothesis says. Again, the fact of China’s low wages is irrelevant to this point, except that low standards of living would normally be a good reason to spend one’s income. This is true for wage-earners. It is also true for recipients of corporate profits.

    Third, if China is receiving more money from its exports than it wishes to spend on imports, it will indeed be running a trade surplus. If this were the result of purely private decisions, there need be no intervention in the foreign currency market by the Chinese government and no creation of Chinese money. Both the intervention and its potential monetary consequences are the product of Chinese government decisions that have nothing to do with some unbridgeable trade surplus.

    The Chinese government stops any purely private capital outflow, through regulation. It has also set an exchange rate at which exports currently exceed imports. And it also follows a set of monetary, fiscal and other policies that sustain the savings (surplus of income over spending) needed to sustain the real exchange rate needed to generate the current account surplus. The rest of the world (particularly the US) adjusts to these decisions. Of course, the reason the Chinese government follows these policies is to sustain the competitiveness of China’s exports at a greater level than would be generated by the market. But that is a policy decision. It is not structural.

    Fourth, it is false to state that trade surpluses and deficits could not happen under the gold standard. Huge and long-standing current account surpluses and deficits were quite normal then (and correspondingly huge creditor and debtor positions), provided people were willing to accumulate claims on assets in other countries.

    Imagine the following simple scenario under the gold standard: the Chinese government buys $250bn a year of US government bonds as a counterpart of fiscal and current account surplus of the same size. Of course, such a policy would have been unlikely under the gold standard.

    Under the classic gold or gold-exchange standard, the monetary consequences of official exchange-market intervention would not have been sterilised. The monetary expansion that would have resulted would have led to inflation, an appreciation of the real exchange rate and higher domestic spending levels. This assumes that Chinese private people did not want to buy $250bn a year of US treasury bonds. If they did want to do that, however, there would have been the same outcome as today: a large current account surplus. But there would have been no inflationary consequences, since the gold that went in via the trade surplus would have gone out via China’s private capital account deficit. The balance of payments would have balanced without monetary effects, despite the trade surplus.

    Finally, the money glut hypothesis fails to explain the persistence of Chinese excess saving in the national accounts. I am prepared to accept that forced savings generated by sterilised intervention, to sustain the trade surplus, is part of the story. But it can only be a part of it. Equally, the money-glut hypothesis fails to explain why the Federal Reserve should follow the policies it does. At the global level, I would return to a previous point I made in this forum: the monetary policy we see is a response to a shift of the IS curve down and to the left – in other words, it is the reaction to global ex ante excess savings. Of course, the monetary policy may be far too expansionary. But it can change only if the excess savings changes, unless we are prepared to accept a global recession.

    I hope this does at least explain why I think Richard’s influential propositions are dangerously wrong, since they are likely to lead to mistaken policy. What is needed, instead, is adjustments in exchange rates, much more spending, relative to income, in China and much less spending, relative to income, in the US.

    Posted by: Martin Wolf | June 26th, 2007 at 9:06 pm | Report this comment
    

15. Richard Duncan: I consider Martin Wolf to be the best financial journalist of our time. Therefore, to my mind, to have him write about me at all is a source of pride, and to have him refer to my propositions as both influential and dangerous (as he does above) is a down right honor. Of course, I would prefer that he agreed with my views, particularly given how well placed he is to further confuse public opinion as to the root causes and potential cures of today’s unprecedented global imbalances if he is won over by the savings glut theorists.

    Martin believes strongly in the benefits of free trade and he devoted almost all of his 16 paragraph critique of my comments to vigorously attacking my opening assertion that freed trade between the Unites States and low-wage countries like China inevitably leads to large trade imbalances on the grounds that it “is in violation of the theory of comparative advantage that has been the cornerstone of economists’ views on trade for two centuries.” In my defense, I would suggest that the theory of comparative advantage was formulated at a time when the gold standard made large and persistent trade imbalances on today’s scale impossible and ages before governments could have even conceived creating billions and hundreds of billions of dollars to maintain their comparative advantage via intervention in currency markets.

    In this century, we find ourselves in a situation of unprecedented global imbalances that are the direct result of, among other things, government actions designed to prevent the adjustment of those imbalances. These are circumstances Adam Smith and David Ricardo could not have imagined in their wildest eighteenth century dreams.

    As interesting as it would be to debate how trade between the United States and China would have played out in a world where the Chinese central bank did not print hundreds of billions of dollars worth of Yuan each year to prevent the Chinese currency and Chinese wages from appreciating, that is not the topic of this particular discussion. In the world in which we live, the PBOC and other central banks did, do and will continue to create money and buy dollars to perpetuate their low wage comparative advantage. The longer influential economists mislead public opinion by diverting attention away from the leading role central banks play in creating global imbalances by assigning blame to workers in developing countries whom they accuse of saving too much out of their meager incomes, the longer this situation will persist and the greater the global imbalances will become.

    In blaming low income earner for not spending enough, the saving glut theory strikes me as not only manifestly wrong and mean-spirited, but dangerously misleading as well. The saving glut theory implies that the global imbalances are the fault of hundreds of millions of individuals in developing countries who stubbornly refuse to spend their sparse earnings and consequently there is nothing that can be done about it. The truth, however, is that the money glut is the direct result of the decisions made by a few hundred central bankers and there is a great deal that can be done about it.

    The question that proponents of the savings glut theory have to answer is this: Is paper money created by central banks “savings”? If the answer is yes, then I agree there is a saving glut. However, I believe it is wrong to call money created by central banks “savings”. Therefore, I see it as a money glut. Call it what you will, take away the $3 trillion dollars created by non-US central banks over the last five years and how much excess savings would there actually be after all? The total amount of tradable US Treasury bills, notes and bonds only amounts to $4 trillion. And, by the way, non-US investors already own half of them. Last year, the foreign exchange reserves of China alone increased by approximately $250 billion as the PBOC printed (not “saved”) the equivalent amount of Yuan. The US budget deficit (and, therefore, the issuance of debt by the US Treasury Department) was less than $200 billion. The PBOC by itself could have bought every new Treasury bond issued last year and still have had plenty left over to buy boodles of securitized sub-prime loans. How surprising can it be then that yields on Treasury bonds remained low despite the Fed’s best efforts to make them rise?

    The global savings glut cannot be understood without understanding the role that non-US central banks are playing in creating that glut. Similarly, any proposals to bring about adjustment in the global imbalances that ignores the role of central banks in creating those imbalances can only fail. That is why it is so important to understand that the savings glut is a money glut.

    So, I ask you, Martin, is money created by central banks “savings”? Is it, Chairman Bernanke?

    Posted by: Richard Duncan | July 2nd, 2007 at 2:18 am | Report this comment
    

16. Martin Wolf: I thank Richard Duncan for his kind remarks about me and for his lengthy reply to my critique. I appreciate it even though he has misrepresented and misunderstood my response in important respects.

    Richard starts by stating that I devoted “almost all” of my 16 paragraph critique to his proposition that free trade between the US and China must be unbalanced. To be precise, my critique of his influential position was in 15 paragraphs, of which the first was an introduction and the last a conclusion. Of the remaining 13 I devoted seven to the trade balance in the absence of currency intervention. But these were relatively short. I actually devoted about half to the real (trade and balance of payments) side and the rest to the monetary and foreign exchange policies of the Chinese authorities.

    It also appears that I should have devoted even longer to the earlier part since he persists with some serious factual and theoretical errors. He seems now to accept (or at least not deny) that it is perfectly possible for cheap-labour countries to trade with expensive-labour countries without running large trade surpluses. Or at least he accepts that they did so in the past. But, he insists, things were different in the old days “when the gold standard made large and persistent trade imbalances on today’s scale impossible and ages before governments could have even conceived creating billions and hundreds of billions of dollars to maintain their comparative advantage via intervention in currency markets.”

    I am going to ignore the remarks above comparative advantage, for reasons of space, since it reveals that Richard is confused between a country’s pattern of comparative advantage in particular sectors (that is, the things it does relatively well) and competitiveness (probably best measured by the overall real exchange rate). The important point, as I have tried to explain, is that huge net capital flows were perfectly possible under the gold standard or, more precisely, the gold exchange standard, which was in place in much of the world in the late 19th and early 20th centuries.

    Today, the absolute sum of global current account deficits and surpluses is about 5 per cent of global product. This is probably not vastly bigger than it was in the 1870s when the UK, the world’s largest capital exporter at that time, ran a persistent net capital outflow (current account surplus) of about 5 per cent of GDP. By 1913, UK net foreign assets were about twice GDP. No significant country yet comes close to these figures (as creditor or debtor) today. Between 1870 and 1889, the net capital inflow into Argentina (that is, the average size of its current account deficit) was 19 per cent of GDP.

    There is, in sum, nothing extraordinary about the size of the current account deficits and surpluses of today. (What is extraordinary, by historical standards, is rather the direction of the net flows, from poor countries to rich ones.)

    This point is important because it is a mistake that runs through Richard’s book. There is, I repeat, nothing impossible about huge current account surpluses and deficits in a gold standard world. All that is needed for this to happen is for private sectors to be happy to accumulate claims on other countries. This is the core of the “savings glut” view, this time set in the 19th century. The British private sector ran a sizeable current account surplus which it invested in the bonds issued by the private and public sectors around the world. This itself required no net monetary transactions (i.e. no net flows of monetary gold) and was entirely sustainable under the gold standard.

    At this stage, Richard brings in the red herring of my supposed criticism of poor Chinese people for saving so much. I have, in fact, gone out of my way in a series of columns to point out that roughly two-thirds of Chinese savings are generated by corporations and the government, not poor individuals, who have good reasons to save as much as they do. But the Chinese government has little reason to save so much in the form of claims on the US treasury, when it could spend more on the Chinese themselves, instead.

    Now we can confront directly the core question Richard asks: is paper money created by central banks savings? This is a misleading question. The paper money created by central banks are not themselves saving. They are money. But savings can be held in the form of money. Remember that one of money’s functions is as a store of value. In a fiat money system, such as today’s, money created by governments is simply a non-interest paying irredeemable government bond.

    While a country does not save when it increases its own money supply, it can save by investing in accumulations of foreign money. This is exactly what China is now doing - irrationally, in my view: it is exporting real goods and services in exchange for accumulations of dollar claims. The accumulations are the counterpart of both the current account surplus and excess domestic savings, as they must be.

    To elucidate what is going on one needs to start from simple assumptions and build up to where we actually are. Imagine that China actually used the dollar as its currency, instead of the renminbi. Imagine, too, that the Chinese saved 50 per cent of GDP, invested 40 per cent of GDP and ran a current account surplus of 10 per cent of GDP. The excess of Chinese exports over imports is paid by foreigners in dollars, which the Chinese then use to buy US treasury bonds. What are the monetary consequences of all this? None whatsoever, is the answer. It is the perfect analogue of the gold standard case and it can go on forever.

    Now imagine that China does not allow its citizens to hold dollars or buy foreign assets with the dollars they earn. Instead they are allowed only to hold domestic currency and bank accounts. Imagine, too, that the Chinese authorities exchange dollars for renminbi at a fixed exchange rate.

    Then the Chinese exporters in the above example must surrender their dollars for renmimbi, which they hold as cash or deposits in the banking system. Meanwhile the Chinese authorities hold dollars, which they again use to buy US treasury bonds. So China’s net asset position is the same as in the previous example, but the Chinese government has a liability to the Chinese public in renminbi that matches its foreign assets and the Chinese public has a (notional) claim on the Chinese government in renminbi .

    Where are we so far? China has a large excess of savings over investment and a large (and matching) current account surplus. China is duly accumulating claims on foreigners and the Chinese public is accumulating monetary claims on the Chinese government. Is the act of money creation itself saving? No, certainly not, but monetary accumulations are the form that a part of Chinese savings takes. If you like, you can say that a part of Chinese savings is hoarded as domestic money itself created as the counterpart of accumulations of foreign assets.

    In passing, let me note that the non-US central banks (or, in this example, China’s central bank) did not create $3 trillion, as Richard seems to suggest. They created the domestic currency equivalent of the $3 trillion. Their countries earned the $3 trillion in trade (and through capital receipts). To the extent that the counterpart was US money, it was created by the US monetary authorities and banking institutions. I suspect Richard and I do not disagree on this point. We also do not disagree, I imagine, that this is an expensive way for the Chinese and others to create domestic money in a fiat money age. Why earn it when you can print it?

    Now let me return to what is going on in China. It is at least conceivable that the Chinese are quite happy to accumulate the renminbi cash and deposits. Remember that the country’s economy is growing at about 15 per cent a year in nominal terms. This, on its own, might justify a growth in the money supply at about that rate. $250bn is roughly 10 per cent of GDP. So if the money multiplier were low enough, the Chinese might find this monetary growth, driven by the external account, perfectly tolerable.

    In practice, it is not. More precisely, the Chinese authorities have had to work quite hard to slow the expansion of bank credit and money, given the rate at which the deposits of the banking system with the People’s Bank of China are tending to grow. So they have to sterilise the latter deposits, by forcing the banks to buy special bonds against which they are forbidden to expand lending. How successful this is and will remain is a controversial question. It has been successful in the medium term, but may not be successful in the long term.

    Now suppose that, as a matter of fact, the overall expansion of the domestic money supply in China exceeds what the Chinese wish to hold. This would lead to asset price inflation and additional spending. In this way, there will begin to be higher inflation, more spending, a diminishing current account surplus and all the rest. The Chinese authorities can let this happen or they can fight still harder to control savings, through higher taxes and so forth. But it is important to be clear that the symptom of their failure would be higher spending and lower savings in China.

    Now imagine yet another possibility. The Chinese government ceases to intervene in the foreign currency market, so letting the renminbi float freely, but frees up exchange controls. Now suppose that the Chinese continue to have domestic savings that are 10 per cent of GDP bigger than investment and a corresponding current account surplus. They then choose to purchase dollars and other foreign assets directly. The exchange rate may not even move from where it is now. The difference from the earlier example is only that the foreign asset accumulation is by Chinese people, not the central bank. There is also still a savings glut, as before.

    So, to bring this part of the discussion to a close, the question whether monetary accumulations or the creation of money are savings is metaphysical and unhelpful. It doesn’t matter what we call things. What we need to understand is the causal mechanisms at work. Is the right model one in which the act of creating money creates the excess savings we see or is it the other way round or is something quite different going on?

    I happen to agree with Richard that the excess savings of the Chinese, shown in the current account surplus, did not just happen. They are in large part the result of policy decisions. But the decisions are not monetary policy decisions – since those are also consequences of the underlying policy decisions. The driving decisions are, in fact, the exchange rate policy.

    The Chinese government has not just a nominal exchange rate target, but a real one. If it had a nominal exchange rate target, as it would have done under the gold standard, it would simply have accepted any inflationary consequences of the net inflow of dollars and consequent domestic monetary growth (see above). But it has, instead, worked hard and successfully (so far) to sterilise the monetary impact and curb inflation. In effect then it is targeting the real exchange rate.

    At this target real exchange rate the Chinese economy tends to generate a huge surplus of exports over imports (now about 10 per cent of GDP). This has both monetary consequences (via the needed intervention in the foreign currency markets) and consequences for spending and incomes. If the Chinese were not to continue to save 10 per cent of GDP more than they spend there would be excess demand, inflation and an erosion of the surplus. In the classic analysis, the external imbalance (i.e. external surplus) would generate a big internal imbalance, too (excess domestic demand). In time, the latter would erode the former. So the Chinese government must sustain the high savings. It can do this via its policy of monetary sterilisation and also, more directly, by saving a great deal itself (as it does).

    So my view is that China’s excess savings (note: not all its savings) has become a consequence of the successful attempt to target the real exchange rate. The interventions in the foreign currency market and growth of dollar assets and domestic monetary liabilities are also a consequence of this policy decision – namely, the decision to nurture export-led growth and a very strong balance of payments. Because China runs a large current account surplus, invested in dollars, as do many other countries, the US runs a large current account deficit, when its economy is in internal balance. The monetary policy needed to keep US spending at about 107 per cent of GDP is also expansionary, for the US and for the world.

    The starting point then is the global real exchange rate configuration and consequent macroeconomic policies. Are low US nominal and real interest rates surprising in this context? No.

    I believe that Richard and I agree that the exchange rate policies are playing a crucial driving role in this story and that both savings and monetary gluts are a consequence, at least in the case of China. In the case of other countries (Japan and the oil exporters) exchange rate intervention per se is not the story. It is the high savings themselves that are the driving force.

    I apologise for this lengthy exposition of some important points. I hope it has clarified issues, even if people do not agree with all elements of it.

    Posted by: Martin Wolf | July 4th, 2007 at 9:25 am | Report this comment
    

17. Richard Duncan: Reading Martin Wolf’s July 4 comment above, it strikes me, now that we have clearly defined our terms, we have reached agreement that the global “savings” glut and the global disequilibrium it has caused are primarily the result of government savings in the countries with trade surpluses rather than the result of savings by individuals in those countries. Let it be noted, of course, that since none of the countries in question have budget surpluses, the “savings” we are discussing are those brought into existence by the central banks of those countries creating money and “saving” it. Now that we can agree on the causes, let’s consider the consequences.

    The debate over the cause of the business cycle has always been the most interesting and the most important in economics. The debate has centred around the question of equilibrium. Specifically, in the absence of government intervention, does equilibrium between savings and investment, and between production and consumption, occur naturally through some mysterious workings of capitalism, or is a destabilizing disequilibrium the capitalist systems’ inherent and fatal flaw?

    In the modern world, however, the assumption of “in the absence of government intervention” is no longer ever tenable. This is important, because our discussion on the savings/money glut has highlighted the role of governments via the creation of savings in causing the current world-wide disequilibrium between savings & investment and production & consumption.

    By implementing government policy (call it an export-led growth policy, an exchange rate policy or a monetary policy, as you prefer), central banks in trade surplus countries have caused the global imbalances and, by doing so, set in motion an extraordinary upswing of the business cycle. Thus far, we have all benefited from the boom stage of this cycle. The point that must be grasped, however, is that in the business cycle, bust always follows boom and big busts follow big booms.

    The Roaring Twenties, the Bubble Economy in Japan and the Asian Miracle were followed by the Great Depression, Japan’s 15 year recession and The Asia Crisis. Each of these booms was marked by an extraordinary build-up in central bank reserves.

    Today, the boom is world-wide. Global foreign exchange reserves have doubled in less than five years. The leading role of central banks in driving this boom via “savings” creation is all too obvious.

    Economists and policy makers need to embrace this fact and begin to consider: 1) how best to rein in the current boom, 2) how to mitigate the coming bust that will be the consequence of this boom and, 3) how to re-establish a rule based international monetary system that will prevent the reoccurrence of the present unprecedented imbalance between savings & investment and production & consumption. Of the three, the second is the most important because today’s global imbalances are so great there is a real risk that a severe global economic downturn with damaging social and political consequences may soon confront us.

    Posted by: Richard Duncan | July 9th, 2007 at 9:44 am | Report this comment
    

18. Martin Wolf: We have got closer to agreement. But we are certainly not there. As I tried to explain quite carefully, central bank intervention is not itself saving. It is the monetary consequence of an exchange rate policy that requires high saving to work. Those savings are, indeed, created by government, directly and indirectly.

    Yet note - this is very important - a budget balance is not itself an indication of whether a government saves, as Richard suggests. Provided a government’s overall fiscal deficit is smaller than its investment it still saves. If, for example, a government has a current budget surplus of 6 per cent of GDP and investment of 10 per cent of GDP, it has an overall deficit of 4 per cent of GDP. But the balance on the current budget is the indication of government saving. On that measure, the Chinese government’s savings are enormous. In 2005 at least, Chinese government savings exceeded the current account surplus.

    Whether a savings glut that takes the form of very large accumulation of foreign assets and large accumulations of domestic monetary liabilities is going to lead to disaster is indeed a big question. I am (somewhat) less apocalyptic than Richard on this. But there are several perfectly plausible solutions: one is generalised floating and so an end to reserve accumulations; another is an end to sterilisation and so acceptance of domestic overheating and inflation in the accumulating countries (this being the classic response to a large increase in reserve money in, say, the gold standard world); another is abolition of controls on capital outflows by individuals, along with greater currency flexibility. I agree with Richard that any of these is likely to be an improvement.

    Posted by: Martin Wolf | July 9th, 2007 at 11:53 am | Report this comment
    

The FT Economists' Forum is a discussion among some of the world's top economists. As a general rule we accept comments from invited members only, but submissions from others will also be considered.

If you are a non-member submitting a comment, please include your relevant academic or financial background.

Post a comment

Name

Mail (will not be published)

As a final step before posting the comment, please type the two words you see in the image beloweight numbers in the audio clip; this test is to prevent automated robots from posting comments.

Type the two words: Type the eight numbers: Unfortunately you provided the wrong reCaptcha answer. Please try again: