October 3, 2006
Why Beijing should dip into China’s corporate piggy bank
China represents something new in the history of the modern world: a developing country that has a vast global impact. This is why Hank Paulson, the US treasury secretary, has followed Robert Zoellick, former deputy secretary of state, in calling for it to be a “responsible stakeholder”. But China will behave as the US wants only if it perceives that this is in its own interests. Again, the US should not be surprised. This is how Americans view their own country’s international obligations.
At present, the most vexed issue between the two countries is the payments “imbalances”. Many in the US complain that China is manipulating its currency, to preserve excessive competitiveness. Certainly, China has a large current account surplus, forecast by the International Monetary Fund at $184bn this year, or 7.2 per cent of gross domestic product. No other country has as big a surplus.
The starting point then must be whether it makes sense for a poor country to export so much capital. The answer, I would argue, is “no”. But we must then also ask why China is running such large surpluses. The short answer is that it is saving even more than it is investing at home. This is true by definition. In China’s case, this surplus is largely being invested in its vast foreign currency reserves, now some $1,000bn (or 40 per cent of gross domestic product).
The remainder of Martin Wolf’s column can be read here (FT.com subscribers only). Discussion from our guest economists is free - click ‘Comments’ below.
Andrew Smithers: I agree with Martin that China should not seek to reduce investment but savings. This is indeed extremely important. Calls for China to reduce investment threaten to make the current imbalances worse. If measures are taken which successfully reduce intended (ex-ante) investment then China’s domestic demand will fall below it optimum level causing an unnecessary and wasteful loss of output both at home and abroad. The accompanying rise in China’s trade surplus would serve to increase protectionism and coming at a time when demand is weakening in the US will increase the risks of world recession.
Reducing China’s savings will not however, occur simply if state enterprises pay increased dividends to the central government. Unless government spending is increased, this would merely transfer the savings from the corporate part of the state sector to the government part. Equally higher taxation of private sector companies will simply transfer savings from the private corporate sector to the government sector.
What is needed is either a cut in personal taxation, which will be at least partly spend on consumption, or an increase in government spending. This could be financed by increased government borrowing, higher dividends from the state owned companies, or higher taxes on the private sector ones, but increased borrowing should be the short term preference, as the other two approaches might reduce investment and by leaving the net balance of domestic savings unchanged, produce no reduction in China’s trade surplus.
The title and emphasis of Martin’s piece should therefore have been different. It should have called for an increase in the state sector’s dissaving, rather than an increase in its expenditure matched by higher revenues.
China’s exchange rate is bound to rise in real terms. As I have pointed out before, when commenting on one of Martin’s previous articles, this is the inevitable consequence of the Balassa-Samuelson effect on the exchange rate of a rapidly growing economy. China can have a rising real exchange rate either by allowing the nominal rate to rise, or by allowing inflation to rise to levels above those of the developed world. The second approach has been implicitly its preferred one, but the method has produced a dangerous time lag. By pegging the currency to the dollar, China has built up huge foreign exchange reserves, which it cannot fully immunise. Left unchecked this will in time produce domestic inflation from over-expansion of domestic liquidity. But there are long and uncertain time lags between increased liquidity and inflation. This raises the risk that when inflation does pick up it will do so in a rapid way which is difficult to control and which will require a sharp tightening of monetary policy and a nasty and unnecessary loss of output to bring inflation under control.
Bringing the inflation impact forward by fiscal stimulus is one way to reduce the risks of delay. I suggest that the title of Martin’s piece should have been “China should either allow the renminbi to rise or use fiscal stimulus to ease global imbalances.”
Posted by: FT Forums | October 4th, 2006 at 12:32 pm | Report this commentAdrian Wood: Interesting information, but a puzzling analysis from as keen a reader as Martin of the work of Wynne Godley. In thinking about the impact of individual sectors (household, corporate, government) on aggregate payments imbalances, it is important, as Wynne has always argued, not to look separately at their savings and investment, but rather to look at the difference between the two, known as the sector’s net acquisition of financial assets (NAFA).
The reason is that the savings and investment decisions of each sector are made simultaneously, not separately. This is particularly true of corporate investment and savings decisions, the world around. Martin may also remember me arguing in my 1975 book, ‘A theory of profits’, that firms’ pricing and profit margin decisions are linked to their saving and investment decisions.
If I have understood Martin’s summary of Jonathan Anderson correctly, the Chinese corporate sector’s NAFA is negative, and is thus not a net contributor to China’s overall payments surplus. The Chinese household sector, I assume, has a positive NAFA, so it still seems correct to pinpoint this sector as the key source of the overall payments surplus.
Martin may well be right that China is investing more than it needs, but that is a different point. I also doubt that the government of China could in practice constrain corporate saving in the way that he suggests – not just for political reasons, but also for the economic reasons explained in my aforementioned book. If you make corporations pay higher dividends, they will raise their prices to cover these, so as to be able to go on generating the level of retained earnings they need to finance their desired levels of investment.
Posted by: FT Economist Forum | October 4th, 2006 at 2:43 pm | Report this commentMartin Wolf and Chinas piggybank
I agree with Martin Wolf that one thing that should be done is that now the state-owned enterprises are profitable, that the Chinese state should use some of these profits. There is no shortage of things that the Chinese state can usefully spend thes…
Posted by: Twofish's Blog | October 4th, 2006 at 6:55 pm | Report this commentFred Bergsten: Martin’s analysis of China both understates the magnitude and urgency of its adjustment problem, at least from an international standpoint, and overstates the difficulty of Chinese action to that end.
Martin cites the standard IMF forecast that China’s current account surplus for 2006 will come in at about $184bn or 7.2 per cent of its GDP. This is clearly a substantial underestimate, based on the trade data for the year to date, and my colleague Nicholas Lardy projects a current account surplus of at least $240bn or 9 per cent of GDP instead. Even more significantly, the juxtaposition of this nominal number with China’s booming growth suggests that its cyclically adjusted surplus is running more like 11 per cent of GDP. These surpluses are growing rapidly and are totally unsustainable from an international standpoint, not least because of the protectionist trade reactions that they are certain to trigger in both the US and Europe once their growth slows and unemployment begins climbing.
Martin suggests a constructive response for China, centred on reducing corporate saving and thus narrowing the domestic underpinnings of the large external surplus. It is not clear that there is as much scope for adjustment through this channel as Martin suggests, however, because the after-tax profits of state owned industrial enterprises in China amounted to only 2.6 per cent of GDP during the first half of this year and 55 per cent of corporate profits accrued to private and other non-state firms, which already pay corporate income taxes and issue some dividends. The Jonathan Anderson estimate for Chinese corporate savings, on which Martin relies, is almost certainly far too high.
On the other hand, Chinese governmental savings are much higher than Martin indicates. They are in fact running at about 8 per cent of GDP when including, as is proper, the government’s budget transfers to state-owned companies that then use those public savings to finance their investments. Martin’s suggestion that the government require the SOEs to pay substantial dividends is sensible but it would be even easier for the government to sharply boost its own spending, especially for badly needed social programs including health care, education and pensions rather than transferring such large sums to the companies.
There is an urgent need for China to undertake economic reforms of both types and I would urge them to include both in their strategy.
Posted by: C. Fred Bergsten | October 4th, 2006 at 9:35 pm | Report this commentWillem Buiter: China saves a staggering amount – its gross national saving rate is just under 50 per cent of GDP in 2005 according to the official data. With gross domestic investment ‘merely’ just over 44 per cent of GDP, the result is an external current account deficit of over 7 per cent of GDP. Even allowing for the almost hilarious inaccuracy of the Chinese national accounts data - the Five Year Plan calls for 10 per cent GDP growth? Ten per cent it will be - there can be little doubt that China saves prodigiously. Why?
Are Chinese equity owners smothered by the corporate veil? Martin argues that one reason for the very high Chinese saving rate is the inability of the owners of Chinese corporations – households and the government - to pierce the corporate veil. When financial markets function well, total private saving is independent of its distribution between household saving and private corporate saving (retained profits). Dividends and retained profits are equivalent sources of income/wealth for households, Corporate profits increase the resources of the owners of the corporation either by being paid out as dividends or by being retained and generating an equivalent amount of capital gains on the owner’s equity.
When a corporation chooses to save more, but households do not want the combined household and corporate saving rate to increase, owners can sell some stock or borrow against the increased value of their equity. This Modigliani-Miller world is not China today. Chinese capital markets are woefully underdeveloped, corporate governance is dreadful, household access to external finance is very limited and there are few liquid household assets that can be sold to undo the impact of excessive corporate saving. In China today, households cannot neutralise, by saving less or by dissaving, the effect on the consolidated private saving rate of an increase in corporate saving.
Martin’s proposition that the same phenomenon may be at work for the state enterprise sector and its notional owners, the Chinese government, is more surprising. Even if it is difficult for the government to force state enterprises to pay dividends, the authorities have the capacity to tax these enterprises, which should take care of any unwanted excessive profit retention by state enterprises. If, for whatever reason, taxing state enterprises is complicated at present, the government could borrow or sell state assets. This would prevent increased saving by state enterprise sector from translating into unwanted increased saving by the consolidated public sector as a whole. The political and state enterprise governance failures that render the Chinese government unable to stop excessive public sector saving are far from obvious. Governments all over the world appear to have a comparative advantage in spending, borrowing and dissaving.
With access to the international financial markets, the option of borrowing is always open to the Chinese state. When would government borrowing in anticipation of future resource transfers from the state enterprises not be possible/make sense? Only if the government believed that it would never be able to get its hands on the retained profits of the state enterprises. Such permanent state enterprise governance failure seems unlikely. So what is different in China? Maybe some fact-finding missions by the Chinese government to Italy, Greece and Hungary would help them discover ways of ‘how to spend it’.
Other reasons for ‘excessive Chinese saving’.
The Chinese social security retirement system is a small-scale, badly designed and poorly administered mess. Coverage is very limited; it is organised and financed mainly at sub-national levels, including municipal and state-enterprise. Private corporate pension schemes are rudimentary at best. Other social safety nets that might put a floor under poverty in old age (including publicly funded health care) are weak or continue to weaken. The increasing private cost of access to adequate health care and reasonable quality education for one’s children (and the uncertainty about the future trends in health and education costs) boosts private saving for both life-cycle and precautionary reasons. Chinese demographics, partly as a result of the one-child policy, has eliminated for most Chinese the expectation/ hope that they will be supported in their old age by their children.
Adequate provision for old age therefore requires private saving. This patterns is reinforced, through a number of channels, by the underdevelopment of Chinese capital markets:
• Annuities are effectively unavailable to Chinese households. This prompts additional precautionary saving to insure against the risk of living longer than expected and running out of assets to finance retirement.
• Financial instruments available to Chinese households, either directly or through institutional investors such as insurance companies and pension funds, offer a poor return compared to what would be available in the global financial markets, and compared to what would be available in the Chinese markets if more internationally competitive providers of saving and retirement products could operate freely in China. For ‘target savers’, a lower risk-adjusted expected return means increased saving.
• Required downpayments for home purchases are high compared to what is the norm in countries with developed mortgage finance markets. Similar constraints affect consumers’ ability to access consumer credit to purchase consumer durables.
Given this long list of reasons why it is individually rational to save a lot, China’s near 50 per cent gross national saving rate becomes less than astonishing. It is therefore pointless merely to exhort the Chinese government to stimulate Chinese consumption. There must be reasons to do so that make sense for China, other than pleasing the US and avoiding a trade war. Three policies would help.
The first would be to create a state pension, administered and financed at the national (central) level, with universal coverage and financed, in the long run, out of general revenues. In the short run, it should be deficit-financed. Any hint of the state pension being related to past contributions should be dropped - this state pension would be fully fiscalised. As long as lower government saving does not lead to an offsetting increase in private saving, that is, as long as the private sector does not fully pierce the government veil, the introduction of this state pension would depress national saving.
The second policy is domestic capital market development. The only way to develop well-functioning domestic markets and institutions rapidly is to radically open up (note: appropriate use of the split infinitive) the Chinese market to entry by foreign financial institutions. Improved corporate governance and better financial supervision and regulation are also key.
The third policy is the easiest from an administrative point of view. It is liberalising capital outflows, at the very least for portfolio investment, but preferably across the board. That is the quickest way to make superior saving instruments available to China’s households, either directly or through yet-to-be created private pension funds and other institutional investors.
Is Chinese investment too low? Living with the Chinese current account surplus.
While it may seem anomalous for capital to flow out of a relatively poor country towards richer countries, I don’t believe that there is much wrong with the Chinese current account surplus. There are good reasons for the very high Chinese saving rates. Tackling the bad reasons will take time, except for the further liberalisation of capital outflows, which could be done overnight.
Given rather slow progress on bringing down China’s saving rate, even given the right policies, the only way to reduce the current account surplus is to boost domestic investment. That would not make sense. China’s 41 per cent of GDP investment rate is already excessive. It is creating an environmental/ecological disaster of unprecedented proportions, first for China and ultimately for the world.
As there are no strong arguments (other than the deeply irrational threat of US protectionism) in favour of China reducing its external surplus anytime soon, nor much prospect of its happening, the counterpart external deficits elsewhere in the world are also here to stay. Their distribution across nations may well change. Fortunately, China is still a quite small share of global GDP: 13.5 per cent in 2005 when measured at Purchasing Power Parity Exchange Rates; less than half that when measured at the, for present purposes more relevant, market exchange rates. It should be possible to live quite comfortably with a Chinese current account surplus equal to 0.5 per cent of global GDP (at market exchange rates). Either way, we will have to live with it for the foreseeable future.
Posted by: Willem Buiter, London School of Economics | October 5th, 2006 at 10:54 am | Report this commentRobert Hunter Wade: Martin Wolf argues that it does not make sense for China to save so much and invest so much, and to build up such large current account surpluses. It does make sense for government to try to cut savings and increase consumption, which would be in the interest of the Chinese people.
Two comments. First on the causes of high savings rates. When I was writing Governing the Market (1990, 2004), about the rise of East Asia, I was struck by how little research attention had been given to the “elephant in the room”, the gigantic savings rates during the fast growth periods of Japan, South Korea and (my focus) Taiwan. Normally hard-nosed economists easily slipped into saying, “Since we can’t explain these high savings rates with anything else, it must be Confucian culture”. On the other hand, it does strike me as plausible that the values and behavioural predispositions around “family” matter for savings rates. Societies where the family is conceived as an entity that stretches far back and far forward in time, for which the present adults are merely the steward, will have higher structural savings - and investment in education - than societies where the family has little duration in time.
Second on the defensive benefits of giant surpluses. In the eyes of the Chinese leadership, the consumption interest of the Chinese people is weighed against, among other things, the national interest of being treated as a global player. Part of the reason for sustaining “irrationally” large surpluses, I suspect, is that other states see them as a kind of atomic bomb. Surpluses of Chinese size induce fear and respect, especially from the US. Invitations to the top table of world finance start arriving. India discovered that the US became more solicitous once it had the A bomb.
This matters to the Chinese leadership especially because it is well aware that once the “war on terror” loses its rallying power over other states, the US will again resurrect the older “fear of nightmares”, namely China as arch-competitor. It will again invoke China as bogey-man, as in the late 1990s. At the moment this strategy is mostly in abeyance because of the awkward fact that China is fully “on board” the US-led war on terror, if only for the legitimacy the war on terror gives to repression on the fringes of its empire. But short of another big terrorist strike on the US mainland, the next US administration will probably look for other, less failure-prone and less expensive ways to exercise a Cold War-type leadership. “China as rival” is the obvious candidate. China’s deterrence will be - in part - its mega-role in the assembly-stages of global production chains dominated by US firms (who will press the administration to moderate its hostility), and the diffuse threat of dollar disruption on an unimaginable scale.
Posted by: FT Forums | October 5th, 2006 at 1:07 pm | Report this commentWillem Buiter: Robert Hunter Wade’s point that “Societies where the family is conceived as an entity that stretches far back and far forward in time, for which the present adults are merely the steward, will have higher structural savings…than societies where the family has little duration in time.” I would argue that extended (over time) families would have the exact opposite effect on saving rates. The dynastic family acts as a family-level unfunded, pay as you go, social security retirement scheme. This reduces any generation’s incentive for saving to fund its retirement. The dramatic fall in the Chinese birth rate under communism (partly due to the one-child policy) has effectively destroyed the dynastic family (from a forward-looking perspective) and has made it necessary for current generations to save for their own retirement rather than relying on the dynastic family’s internal social security retirement scheme. This has contributed to the very high Chinese private saving rate.
Posted by: Willem Buiter, London School of Economics | October 5th, 2006 at 1:37 pm | Report this commentMartin Wolf: Before reacting to these interesting (and varied) comments, let me note that the UBS numbers were slightly in error. A corrected version has been posted on ft.com . The text of my column has also been slightly altered as a result.
The points made in my piece are, first, Chinese gross savings are astonishingly high; second, household savings account for only about third of such savings, the rest being corporate or government savings, with the former being much the more important of the latter two; third, these overall savings are higher than makes sense for China itself; fourth, China’s investment rate is excessive, as well, though cutting that without addressing the savings rate would tend to increase the country’s already large current account surplus; and, finally, one way of reducing the aggregate savings rate is for the Chinese government to take dividends from the companies it owns and then spend them. Maybe this was too much for one column! But I will take each point in turn below.
On the first point, everybody agrees: China’s savings rate is astonishingly high, so high, in fact, that the country runs a very large current account surplus despite what is also an almost equally astonishing investment rate.
On the second, point, there is little disagreement that households are directly responsible for only about a third of the country’s savings. There is disagreement on the relative size of corporate and government savings. Fred Bergsten, following Nick Lardy, the respected China expert at the Institute for International Economics, argues that government savings are far higher and corporate savings lower than Jonathan Anderson of UBS suggests. I am not an expert on the details of the Chinese national accounts. I relied, in this case, on Mr Anderson. But this debate, though interesting, is not really important for policy. If government savings are that much higher, then it is even easier to argue that it both can – and should - spend more than it now does.
On the third point, that China’s savings are excessive, there is considerably less agreement. This is partly because there are two different points here. One, strongly advanced by Fred Bergsten is that China’s savings surplus – the excess of its savings over its domestic investment – is globally disruptive. This relates to the issues that we debated on this forum last week (see the debate on “A slowing US could brake the world”, my column of September 26th).
I would add to this debate that the real return on China’s current foreign assets (after allowing for likely changes in the nominal exchange rate and domestic price level) is almost certain to end up negative. I would also argue, against Robert Wade, that any prudential or “great power” justification for continuing with current rates of accumulation of foreign currency reserves is unpersuasive. The latter are now big enough to cope with any conceivable external shock. Continued accumulations are more likely to strain China’s international relations than strengthen then, while the threat to dump the reserves is hardly credible, since China itself would suffer huge losses. The line of argument is too mercantilist for my comfort.
The other argument that savings are excessive is that China is postponing too much of its consumption. I agree with Willem Buiter that there are good reasons why Chinese households would save a great deal. I agree that it makes sense for Chinese households to invest some proportion of their wealth abroad. I agree, too, that the Chinese government should allow them to make such investments: its current monopoly of investment abroad cannot be efficient.
Yet current corporate and government savings are not part of a rational response by Chinese households. To argue otherwise would be to take the assumption that households are able to penetrate the corporate and the government veils to excessive lengths. In other words, neither Modigliani-Miller (for Chinese corporations) nor Ricardian equivalence - the proposition that the private sector fully offsets government savings - holds in China. Moreover, the government cannot easily lower household savings by any direct measures (though radical economic and financial reform might ultimately achieve that result). But it is perfectly possible for the government to lower its own savings or those of the corporations it owns.
There are powerful reasons for the government to do just this. Public and private consumption are now only half of GDP. Some essential public services have collapsed, while hundreds of millions of people live in great poverty. Surely, an increase in government current spending is justified. Furthermore, as Willem points out, some of the reforms China should introduce, such as a modest universal pension, would raise spending now and depress both government - and, quite possibly - household savings.
Adrian Wood takes a slightly different view on this. Corporations have a financial deficit (i.e. they invest more than they save), he notes, while households have a surplus (i.e. they save more than they invest). So it is Chinese households, he argues, that are responsible for the excess savings that show up in the current account surplus.
I don’t think this matters for two reasons: first, it is perfectly normal in a fast-growing developing country for household savings to be transferred to the corporate sector for investment; second, the Chinese government can more easily alter its own savings and those of the corporations it owns. Adrian responds by saying that the latter can readily raise their prices and so recover what the government may take in dividends or taxes. I am not convinced that this would be easy in an internationally open economy with increasingly competitive product markets. Alternatively, he suggests, corporations would cut investment. I don’t see why they should do that either. Corporations could just borrow a bit more from households, via the commercial banks (instead of letting the same funds be lent abroad, via the banks and the central bank).
On the fourth point, that China’s domestic investment is excessive, there is also some disagreement. Certainly, China’s domestic investment rate is very high even by the standards of other fast-growing east Asian countries – such as Japan and South Korea. But if the investment rate were to be cut, saving would need to be cut at least as much (if not more). Otherwise, the current account surplus would rise still further, as I note above. I cannot see the world tolerating a Chinese surplus of $300bn-$400bn (though one could just thank the Chinese for their generosity, instead).
On the final point, I agree entirely with Willem and Andrew that taking dividends from the corporations it owns is just one way to lower the sum of corporate and government savings. The government could reduce its own fiscal surplus by cutting taxes and/or raising spending. I intend to turn to this in subsequent columns.
Posted by: FT Forum - Martin Wolf | October 5th, 2006 at 6:14 pm | Report this commentFred Bergsten: Martin Wolf makes two additional points on the Chinese currency issue that deserve brief elaboration.
First, he notes correctly that the Chinese will eventually take very large capital losses on the dollar holdings that they have built up to maintain their undervalued currency. The problem is that the Chinese do not, and presumably will never, mark their reserves to market and thus realize those losses. There is no domestic accountability on the issue and hence no tangible cost to the government for investing so much of the nation’s resources in such an imprudent manner.
The unhappy truth is that the Chinese view their currency intervention as an off-budget export and job subsidy that escapes all domestic accountability and, at least to date, any serious international reaction. Even if some officials do care about the inevitable capital losses that will ensue, the government as a whole finds currency undervaluation to be a uniquely attractive policy tool and has already implicitly written off those losses as well worth incurring.
Second, Martin is also right that there is no credibility in the “threat” that China might dump its dollar reserves. It will certainly not sell them for renminbi as this would drive up the exchange rate that they are so intent on undervaluing. More important than his point about the self-inflicted losses that would result, moreover, is the global opprobrium that China will bring on itself by such a disruptive move. The Europeans would of course be mortified by any Chinese shift from dollars to euros as it would drive up the exchange rate of the euro against the dollar. China seeks increasing international approval and respect on international financial issues, as indicated by its behavior throughout the Asian financial crisis, and will hardly throw that away through policies that would either drive up the value of its own currency or anger other key countries.
Posted by: C. Fred Bergsten | October 6th, 2006 at 8:30 pm | Report this commentMartin Wolf: I agree largely with Fred on these points, though I still wonder whether there is not some limit to the willingness of the Chinese government to accumulate foreign exchange reserves. At the present rate, these will reach $2,000bn in four years. At some point, surely, enough will be enough.
Posted by: FT Forum - Martin Wolf | October 8th, 2006 at 6:57 pm | Report this commentWillem Buiter: Re: Fred Bergsten’s assertion that “The Europeans would of course be mortified by any Chinese shift from dollars to euros as it would drive up the exchange rate of the euro against the dollar.” They (the Eurozone inhabitants) should and would be delighted rather than mortified. If they don’t like a stronger euro, all they have do is supply the extra euros the Chinese authorities are demanding at the given euro-dollar exchange rate.
Is Fred asserting that the ECB (the Eurozone agency that should supply the euros, following a portfolio shift by the Chinese international reserve managers out of US dollars and into euros), would not be able to figure out what the Chinese authorities had been up to? Or that, having figured it out, the ECB would not be willing or able to give effective operational expression to the concept of ‘printing euros’, even if doing so would be without prejudice to their primary price stability target? I am second to none as regard my (marginal and average) propensity to attribute mental leaden-footedness to the ECB, but I am sure they would be able to figure this one out.
Posted by: Willem Buiter, London School of Economics | October 9th, 2006 at 12:38 am | Report this commentMartin Wolf: Willem is right or at least I hope he is.
Posted by: FT Forum - Martin Wolf | October 9th, 2006 at 12:28 pm | Report this comment