Consumption and production

You are currently browsing the archive for the Consumption and production category.

The IMF warns about surplus countries and global imbalances

October 3, 2009 in Asian development model, Consumption and production, NPLs by Michael Pettis | No comments

\

\

As Beijing slowly unlocks from its 60^th anniversary celebrations \’96 the streets are still relatively empty but more and more people are going out, although my local Starbucks still hasn\’92t reopened, forcing me to go elsewhere for my hardcore caffeine fix \’96 a lot is still going on in the rest of the world. Both the US and the IMF have come out with releases that help us to pick through the problems that China and the world are facing.

Before discussing these releases, let me make a quick digression to an event that a lot of people have been asking me about. Two weeks ago China Construction Bank announced that it would rollover 24.7 billion yuan in bonds that it had \’93purchased\’94 from its AMC, Cinda, for another 10 years. Bank of China and ICBC, which sit on 473 billion yuan worth of AMC bonds, will probably do the same when their AMC bonds come due.

What does this all mean? Remember that as part of the recapitalization of the banks after the NPL fiasco of 10-15 years ago, the AMCs (asset management companies) were created to purchase and liquidate the bad debt. There is a big argument as to whether or not they took out all the garbage loans, but at any rate they bought a lot of bad debt and, since they had no assets of their own, paid for them with issues of medium term bonds, which they exchanged in two tranches. One tranche was for 100% of the face value of one portion of the bad loans they took on, and the other was for 50% of face of the rest of the bad loans they acquired.

The problem of course is that these bad loans were worth a lot less than either 100% of face or even 50% of face. In fact they have been liquidated at a rate of about 20% of face. This leaves the AMCs bankrupt and unable to repay the bonds, so when they came due the bonds were simply rolled over. There is a sort of comfort letter from the Ministry of Finance (its exact value is in dispute), so the banks have been able to get away with treating the bonds as money good. The point of all this is to remind us that all the .losses for the earlier spate of bad loans, even assuming that all the bad loans were identified and cleaned up (which I doubt) have not been resolved.

Someone (the banks? The Ministry of Finance?) will eventually have to pay up. If the process is allowed to drag on for many years, I suspect that the banks will pay out of retained earnings, but since retained earnings at the banks consist primarily of the very wide spread between the lending rates and the interest rates that banks are allowed to pay depositors, ultimately this means that households will be forced to recapitalize the banks. If there is a short term problem, however, perhaps leading to a crisis of confidence in the banks, I suspect that the MoF (unless debt at the sovereign level in the mean time becomes a problem) will explicitly guarantee the bonds or take them directly on the government balance sheet.

US unemployment picture is ugly

To return to the rest of the world, unemployment in the US is not getting better. Yesterday the Labor Department released figures that showed the US unemployment rate climbing to a fresh 26-year high of 9.8% in September. According to an article in the Financial Times:

Official figures on Friday showed that non-farm payrolls dropped by 263,000, making it the 21st consecutive month that the US economy has shed jobs. The data were worse than even the most grim expectations, as economists predicted a 175,000 drop in payrolls, and followed a decline of a revised 201,000 jobs in August when the unemployment rate was 9.7 per cent.

Although I think most economists are expecting that US economic growth in the third quarter was a fairly healthy 3%, as far as China is concerned it is not the future growth in the US economy that matters so much as future growth in US consumption. A jobless recovery in the US, if that is what we get, probably means that dragging household consumption will not be the engine of US growth, and even less will it be the engine of Asian growth, which it was for so many years. Any Asian and Chinese recovery predicated on a revival of out-of-control US consumption is likely to be disappointed.

On Thursday the IMF released its World Economic Outlook, which was mildly positive on the global economy, arguing that \’93the recovery has started, financial markets are healing, and in most countries growth will be positive for the rest of the year as well as in 2010,\’94 although in line with the US employment report it worried that \’93the pace of recovery is expected to be slow and, for quite some time, insufficient to decrease unemployment\’94 (later in the report they say \’93the current rebound will be sluggish, credit constrained, and, for quite some time, jobless\’94). The report also argued that because most of the \’93recovery\’94 has been based on public spending and, I guess especially in Asia, gearing up capacity without much regard for demand, an economic recovery was likely to be slow and risky.

The IMF seems increasingly to be agreeing with the \’93global imbalances\’94 analysis of the economy, probably to the dismay of China and other surplus countries. Early in the report it says:

To complement efforts to repair the supply side of economies, there must also be adjustments in the pattern of global demand in order to sustain a strong recovery. Specifically, many economies that have followed export-led growth strategies and have run current account surpluses will need to rely more on domestic demand and imports.

Surplus countries must consume more

The interesting thing for me was this focus on surplus countries. Although there does seem to be an economic rebound, the report says, the recovery will be weak unless countries with large trade surpluses step up domestic demand. To keep growth up, surplus countries like China must boost domestic spending, and appreciate their currencies. This pretty tough claim will probably not make Beijing, Berlin or Tokyo very happy, although it does chime with US views on global trade imbalances. In their own words:

This will help offset subdued domestic demand in economies that have typically run current account deficits and have experienced asset price (stock or housing) busts, including the United States, the United Kingdom, parts of the euro area, and many emerging European economies. In these economies, private consumption and investment are unlikely to pick up the slack that will be left by diminishing fiscal stimulus, given that household incomes and corporate profits will be subdued and balance sheet repair will be under way for some time, implying higher saving rates.

The authors of the report do not seem terribly optimistic about the prospects for a sustainable spurt in surplus-country domestic demand in the near term (\’93This process of rebalancing global demand will be drawn out.\’94) but I am not sure, perhaps because the IMF is after all a very politicized institution, that they specify the trade consequences. They acknowledge that there will be a problem with expected increases in savings in one part of the world conflicting with high savings elsewhere, and they don\’92t seem very optimistic about prospects for a surge in investment, but it seems to me that they shy away from working out how this will happen and how the pain will be distributed (through the trade account, I would argue).

What about overinvestment?

In a section in Chapter 4 of the report entitled \’93Do Precrisis Conditions Help to Predict Medium-term Output Losses?\’94 there was an interesting discussion about the relationship between output losses associated with a crisis and pre-crisis investment levels. On especially commented on section had this:

The prominent role of investment and capital losses suggests that the level and evolution of precrisis investment would be good predictors of eventual output losses. Indeed, regression results provide strong evidence that economies with high precrisis investment-to-GDP ratios, measured as the average investment-to-GDP ratio during the three years before the crisis, tend to have large output losses.

In contrast, the investment gap, defined as the deviation from its historical average of the investment-to-GDP ratio during the three years before a crisis, is not statisti\’adcally significant. We return to potential interpretations of these results later in this section, but it is worth mentioning that the precrisis investment share is particularly robust as a leading indica\’adtor, even after controlling for the level of the current account balance. This suggests that countries that have high investment rates tend to experience larger output declines follow\’ading banking crises, irrespective of whether the investment is financed by foreign or domestic savings.

For those of us who worry about China\’92s having recently increased its already-excessively-high investment rate, this passage was an uncomfortable read. In addition for people like me, who believe strongly that the very process of misallocated investment will act as a damper on future consumption growth (and I think this is becoming much more widely accepted, or at least discussed, in policy circles), the combination of warnings over overinvestment and pleas for more consumption from trade surplus countries is deeply worrying. By the way, for a short and quick view of why I think consumption won\’92t grow, you can check a recent debate held by the New York Times on the subject of Chinese consumption growth.

So what about all this excess investment? The State Council recently made a lot of noise about its determination to curb excess capacity. Here is the Financial Times version of the story:

China has issued a stark warning about the risk from rising overcapacity in the economy, saying it could hamper recovery and lead to a surge in non-performing bank loans. The State Council, the country\’92s cabinet, issued a new plan to combat overcapacity in seven industries, barring new aluminium smelters for three years and criticising \’93blind expansion\’94 in parts of the steel and cement industries.

The cabinet statement, which came late on Tuesday evening in Beijing, follows a crescendo of warnings from senior officials. It also outlined measures to restrict manufacturing of equipment for \’93green\’94 industries of wind and solar power. China\’92s economy has rebounded sharply in recent months due to an investment boom \’96 much into infrastructure \’96 fuelled by increased public spending and a surge in lending by the state-owned banks.

But over the past three months many government officials have begun to publicly warn that the credit binge could create overcapacity in heavy industry, which could produce a new round of bad bank loans.

The article in the South China Morning Post adds some color, and a partial explanation of why all these angry statements about preventing excess capacity over the past few years have had so little effect:

In unusually blunt wording, the cabinet also pointed its finger at local authorities. \’93Some regions have acted illegally. We are once again seeing cases of illegitimate approvals, of construction starting before it has been approved, and of construction starting even as the approval process is underway,\’94 it said.

The cabinet\’92s strident warning about overcapacity underscored why officials have been circumspect about the economy, repeatedly saying that it has shown signs of recovering from the global financial crisis but is still not on solid ground.

It is hard to give up investing

The truth is everyone in the world is against the creation of \’93excess\’94 capacity, but as long as Beijing has in place policies that explicitly subsidize investment and production, it will take an awful low more than fulminating against wasteful investment to eliminate it. I would argue that wasteful investment is the automatic consequence of policies that lower the cost of capital to \’93unreasonable\’94 levels, implicitly socialize risk, and otherwise subsidize producers in the name of boosting employment.

Since Beijing has very explicitly chosen to attack rising unemployment in the short term \’96 probably wisely, although also probably more ferociously than was optimal \’96 there is little they can do to prevent a massive rise in wasteful investment. You cannot take an economy with the highest investment rate in history, and already massive waste, and very quickly force investment rates up even higher, without also increasing waste. The problem with all this wasted investment, of course, is that someone must pay for it, and that \’93someone\’94 will undoubtedly be Chinese households, who will then almost certainly go on to disappoint us by failing to splurge on consumption.

And are they really serious about tackling excess capacity? Here is what Bloomberg said in an article earlier this week about the shipping industry:

China and South Korea\’92s support for shipbuilders may add to a glut of capacity, slowing a recovery in freight rates and vessel prices. The world\’92s two largest shipbuilding nations have taken steps this year to aid shipyards and safeguard jobs as customers delay or scrap orders amid tumbling world trade. That support will likely ensure more vessels enter service, even as lines mothball and scrap existing ships because of a lack of cargo.

\’93The Chinese and Koreans, in particular, will make sure that these ships come,\’94 Philip Clausius, chief executive officer of lessor First Ship Lease Trust, told a conference in Singapore yesterday. The \’93daunting number\’94 of ships that \’93will hit the market over the next three, four, five years will make the recovery a rather slow and painful one.\’94

China\’92s bid to become the largest shipbuilding nation by 2015 may also worsen the glut as it competes for market share, said Matthias Umlauf, senior economist at HSH Nordbank AG. The world\’92s shipyards have dry-bulk ship orders with a combined capacity of 64 percent of the existing fleet, according to data compiled by Bloomberg.

China has \’93the chance to become the world\’92s largest shipbuilding nation and they will not let this chance go,\’94 said Umlauf. \’93They will support their national champions and that will definitely add to the overcapacity situation.\’94

As I have said many times before, I don\’92t see how pressures to increase savings in the US and other trade-deficit countries will not conflict with pressures in China, Germany, and other trade-surplus countries to maintain policies that force up savings rates, especially if sustainable global investment rates decline. The only outcome, I think, is increasing trade tensions. In that light, today Bloomberg reported a very worrying escalation of the conflict:

The two largest groups representing U.S. companies in China said the Asian nation has enacted a series of policies discriminating against foreign investors and imports. The U.S. Chamber of Commerce and the U.S.-China Business Council said in testimony today that Chinese contracting rules, technical standards and licensing requirements were protectionist. Chinese officials have made the same charge against the U.S. following President Barack Obama\’92s imposition of tariffs on Chinese tire imports.

Both organizations have previously defended China, calling it a large and growing market for U.S. exports and lobbying to fend off legislation aimed at punishing China for currency policies and government subsidies. The criticisms of the two U.S. groups reflect mounting tensions that economists said could spark a spiral of retaliatory measures between the countries.

\’93There are growing indications that China\’92s movement toward a market economy has stalled,\’94 Jeremie Waterman, senior director for China at the U.S. Chamber of Commerce, testified to a hearing at the U.S. Trade Representative\’92s office today. \’93The voices of protectionism in both countries are on the rise.\’94

More trade tensions, and the very limited advantage of relative poverty

September 25, 2009 in Consumption and production, Exports and imports, Trade protection by Michael Pettis | No comments

While the G20 leaders make reassuring noises about international trade, I think the risk of rising trade tensions have not abated at all. As I see it, everything depends on whether or not domestic Chinese polices had any role in creating the global imbalances, and if they did, then we are still in the early stages of a difficult process of assigning the costs of the global adjustment through trade.

Beijing hates when anyone suggests that Chinese policies were partly at fault for the current global imbalances, and doesn\’92t even like people to use the phrase \’93global imbalances,\’94 but like it or not, we have to figure out whether in fact Chinese policies mattered. As I see it, China\’92s consumption rate, the lowest ever recorded, and it\’92s trade surplus, the largest as a share of global GDP ever recorded, could not help but have been caused by policies \’96 such as an undervalued currency regime, excessively low interest rates, sluggish wage growth, unraveling social safety nets, and manufacturing subsidies \’96 that were almost wholly under domestic control.

According to my understanding of Chinese growth, it was policies that systematically forced households implicitly and explicitly to subsidize often-otherwise-unprofitable investment and manufacturing that led to wide and divergent growth rates between production and consumption, and of course the gap between the two is the savings rate. If that is true, the stimulus package is only likely to exacerbate the domestic imbalance.

This matters because as the US begins the too-slow but irresistible process of raising its savings rate, something else must change too. At the global level savings must of course balance with investment, and with general expectations that investment will at best remain steady and probably actually decline over the next few, a rising US savings rate must result in one or more of three outcomes:

1. Total US savings do not rise \’96 which means US GDP must contract as the savings rate rises

2. The savings rate in the rest of the world declines, or at least grows much more slowly than in the past. Since China is the country with the highest savings rate and the largest trade surplus, this means China\’92s savings rate will decline, and this is just another way of saying that consumption growth will surge.

3. China\’92s GDP grows much more slowly.

So we are left with the almost inescapable fact that if the US savings rate increases, either China (and the rest of the world, technically, but in practice mainly China) must see much faster consumption growth or the world must experience a slowdown in GDP growth.

Consumption growth determines trade tensions

How quickly can China raise its consumption growth rate? Optimists, and those who think that Beijing\’92s policies did not contribute to the global imbalances, believe that the fiscal and credit expansion of the past several months can cause both investment-led growth and a sustainable rise in consumption growth. Pessimists point out that it was exactly these sorts of highly inefficient investment-driven policies that left China with its savings and trade imbalances, so that intensifying them can only exacerbate the imbalances over the medium term.

If the optimists are right, and China sees a long-term and sustainable surge in consumption, most of the brunt of the global adjustment will take place in the US, and China and the rest of the world will return relatively quickly to growth. If the pessimists are right, and of course I am a pessimist, the global economy is likely to suffer a period of struggling growth as tendencies to force up global savings conflict with the tendency of global investment to decline.

In that case the main mechanism for distributing slower growth among the world\’92s major economies will be through international trade. Differences in the savings and investment rates in each country show up as surpluses and deficits in the trade and capital accounts. With consumption being the most valuable commodity, both trade surplus countries, with their consumption deficits, and trade deficit countries, with their consumption surpluses, will be maneuvering ferociously to access as much global consumption as they can. In that case expect a sharp and continuing rise in trade tensions. The G20\’92s best intentions won\’92t matter.

This, by the way, seems to be a repeat of the Japanese story in the 1980s and the 1990s. As regular readers of my blog know, I believe there are lessons for China from what happened to Japan after the US stock market crash in 1987 signaled the need to end Japan\’92s dependence on a burgeoning US trade deficit to absorb its excess capacity. Japan then, as China now, responded to the collapse in its biggest export market with a credit and fiscal expansion that at first protected Japan from the employment consequences of the contraction in US net consumption, but which ultimately may have exacerbated Japan\’92s imbalances and made its adjustment all the more difficult.

The Japanese parallel

I\’92ve been speaking to a lot of investor groups in the past month, and when I discuss the parallels between China today and Japan after the 1987 US stock market crash I am often told that the comparison isn\’92t useful because of one (or both) of two major differences. The first is that since China\’92s current consumption level is so much lower than Japan\’92s in 1987, it is far more reasonable to expect a surge in Chinese consumption to replace the declining US demand for Chinese excess capacity than for a surge in Japanese consumption to have done the same after 1987. Japan might not have been able to pull it off, but, they say, it is much easier for China to do so because it is so much poorer and starting from a much lower base.

The second objection \’96 perhaps not so different from the first \’96 is that since China is so much less developed than Japan was in 1987, an infrastructure investment surge is a lot more sustainable. After all, Japan already had great infrastructure in place at the time, so that much of its new investment after 1987 was inevitably in the form of highly wasteful \’93bridges to nowhere\’94. Since China has much lower quality infrastructure stock, they argue, there is much more it can do in the way of sustainable investment.

I am always a bit puzzled by how widely-held these views seem to be, especially in China but also abroad. The idea that being poorer makes policy easier can\’92t have emerged from looking at the experience of developing countries. I suspect that it arises from assuming that poverty does not represent differences in real factors \’96 worker productivity, education, the institutional and legal framework, etc. \’96 so much as in policy mixes.

It is true that poorer countries are able generally to achieve faster growth rates than richer countries, perhaps because they have only to play catch-up, but there is little evidence from other countries that poverty leads systematically to more profitable investment or to more sustainable consumption growth. I think both objections stem from implicit assumptions that there is some highly attractive upward limit to either consumption or infrastructure investment, and that the further away we are from that limit the stronger the attraction towards it. But if that assumption weren\’92t mistaken poverty should have ended long ago.

Take consumption. At the very least if consumption growth were an inverse function of wealth, or of existing consumption levels, the US would have the slowest consumption growth rate in the world and certain African or Caribbean nations would have the fastest. This clearly isn\’92t the case.

Household income growth determines consumption growth

I would argue instead that the growth rate in consumption is partly a function of demographics and income distribution, partly a function of the willingness of banks to increase or reduce consumer credit, and more generally a function of the growth rate of household income. Other things matter too \’96 for example I agree with many of my colleagues in and out of China that a good health insurance system may reduce the need for Chinese households to save since it smoothes out expected health costs \’96 but it seems to me that absolute level of wealth is almost irrelevant in determining potential consumption growth rates. Rich people, after all, seem as determined to increase their consumption as poor people (you can easily see that in the behavior of the hordes of the new wealthy in Beijing and Shanghai), although of course the goods and services they will want to buy will be very different.

In that case what really matters to Chinese consumption growth is the rate at which wages and other forms of household income grow, and the extent of implicit taxes or subsidies that penalize or favor consumption. I exclude possible growth in consumer credit because Chinese banks have never figured out how to do this without a rapid increase in non-performing loans.

Of course it is very important to remember that household income in China is not just wages. Interest on bank saving deposits is also an important source of income, as are various social transfers. There are also a variety of hidden taxes on household income \’96 some obvious and very significant, like the low deposit rates the PBoC demands to subsidize bad lending practices and otherwise non-viable investments, others less so, like an undervalued exchange rate, which effectively creates a consumption \’93tax\’94 on imported goods.

These are the things that matter. While other factors may affect consumption rates at the margin, I think it is pretty clear that the growth in total household income \’96 wages, interest income, and other social transfers including the various \’93safety nets\’94 \’96 largely determine the growth rate in consumption in China, Japan, and in almost any country. If this is true, the relative wealth or poverty of a county says little about future consumption growth, and the fact that China is much poorer today than Japan in 1987 in no way should convince us that it will be that much easier to boost Chinese consumption.

Two asides

It is worth making two asides which may seem obvious, but are often lost in discussion. First, in discussing the resolution of global imbalances we need to take gross amounts into consideration. In other words because both the Japanese and the US economies are so much larger than China\’92s, and their consumption rates higher (more than twice as high, in the case of the US), a 1% slowdown in US consumption is not dissipated by a 1% growth in Chinese consumption, and a 1% increase in Japanese consumption does not have the same effect as a 1% increase in Chinese consumption. In both cases the change in Chinese consumption would have to be much greater.

Second, there is a big difference between consumption growth and growth in the consumption share of GDP, and this difference matters very much to the whole rebalancing debate. If Chinese consumption is growing at the 8-9% rate characteristic of the past several years, it still might not resolve the problem of a decline in US consumption even though by any standard that would represent a rapid rate of growth. If Chinese GDP is growing faster than this, as it has done for the same period, the imbalance is not only not being resolved, it is getting worse.

Chinese consumption, in other words, has to grow faster than Chinese production over the medium term in order replace a decline in net US consumption. High growth rates in China do not resolve the imbalance if production grows faster than consumption.

This is a very long way of saying that in comparing of policy responses the lower level of consumption in China is not at all an important difference between China today and Japan in 1987. Even if it creates more \’93room\’94 for a rise in Chinese consumption than in Japanese consumption \’96 a claim about which I am very skeptical \’96 it does not make it any easier for Chinese consumption to rise to the challenge in a way that Japan could not. It still means very broadly that over the medium term Chinese household income will have to rise faster than Chinese GDP \’96 something it has not been able to do at all in the last decade \’96 in order for China to absorb the declining net demand from the US for Chinese goods once its government-fueled investment boom peters out.

But what about investment \’96 must the government-fueled investment boom peter out? China has a much weaker and lower quality infrastructure than Japan did in 1987, so it seems a safe bet that China can sustain its investment boom for a lot longer than Japan could, right? This is the second objection to the comparison between China today and Japan in 1987.

Again, I think this is a fallacy. Let\’92s leave aside the obvious problem that much of China\’92s infrastructure investment may be wasted on spending that has no social benefit or simply is stolen, not because this is a small problem but rather because most of us would easily understand that a government debt-fueled investment boom to finance the purchase of private homes in Paris or Los Angeles or even large swimming pools and luxurious dining facilities for local municipal officials must still be repaid, and that it will be repaid out of future household income that would be better and more fairly spent on future household consumption.

The problem is that even \’93good\’94 infrastructure projects, like airports, railroads and highways, also have limits. These projects have to repay their cost, including the appropriate cost of capital, because if they don\’92t, the payment must anyway be made out of future household income, acting as a drain on future consumption. Some projects can pay for themselves, and some might not pay for themselves directly but can increase economic value so that ultimately, by creating wealth, they effectively pay for themselves out of higher future income. In either case households are left wealthier even after paying for those projects, and so able to consume more.

Productivity matters

But does relative poverty really improve the value of these investments? It might seem obvious that taking a good railroad system in Japan and turning it into a state-of-the-art railroad system increases the value of the railroad less than taking a bad or non-existent railroad system in China and turning it into the same state-of-the-art railroad system. In that case China seems to have more scope for additional investment than Japan does.

But does it? Maybe not. Japanese labor costs a lot more than Chinese labor, and is far more productive, so it is not clear that the improvement in labor efficiency caused by the railroad investment is necessarily more valuable in China than in Japan, even though the absolute change in quality of the railroad service in China is certainly higher than in Japan in my example.

That is I think core of the problem. The scope for nominal improvement in infrastructure is certainly higher in China than in Japan, but nominal improvement doesn\’92t matter. It is the economic value of that improvement that matters, and the economic value of improving the railroad in China is not necessarily higher than in Japan since, for example, every hour of transportation time saved in Japan may be substantially more valuable than an hour saved in China.

In fact I would argue \’96 as have many economists, by the way \’96 that China\’92s obsession with high-technology or state-of-the art infrastructure is extremely wasteful because the benefits of the most advanced technology only justify the costs if labor productivity and labor costs are very high. This is perhaps another way of saying that China\’92s highly capital-intensive growth is far from optimal for China, and probably only reflects the fact that capital is so cheap in China, at least for the capital-intensive SOEs that get the bulk of bank financing. This means that achieving Japan-style levels of infrastructure are not necessarily the best way to invest in infrastructure. The optimal infrastructure level in China is lower than the optimal in Japan, so the fact that China starts from a lower base does not automatically mean that it has more scope for profitable investments.

Airports are perhaps a good way of thinking about this. China doesn\’92t have as many airports as Japan does (adjusting for size and population), so clearly that means that China can engage in an airport-building spree that would be folly in Japan, right? Maybe not. Chinese are far less likely to be able to afford air travel than Japanese, and are less likely to need to ship goods by air than are the Japanese, so China needs efficient air travel much less than does Japan. Simply pointing to the fact that China has fewer airports does not imply that it has more room to build airports. In fact in my opinion it is very likely that we are going see so much money spent on Chinese airports in the next few years that it is almost impossible that we will ever recoup their cost.

As an aside I am often told about, as another example of the kind of investment spending that can pull China out of the crisis, the building of \’93shadow\’94 cities next to older ones, with much better facilities. Eventually everyone is expected to move out of the old city, with its less than optimal facilities, to the new state-of-the-art version. If enough cities do this, the argument goes, China can achieve huge growth rates.

Of course it can, in the short term. And if the US government were to raze Chicago and immediately rebuild it, I suppose that they could build a far more efficient city and would certainly create a huge short-term boost to the local economy (for one thing they would probably wipe out local unemployment).

Spending must be justified

But is this a good idea? If the US government were to propose doing it I am sure President Obama would meet with a storm of criticism. It would be pointed out that the increase in productivity created by this new, improved Chicago would almost certainly be only a fraction of the cost of rebuilding the city, and the difference would represent a straight increase in net indebtedness.

They would almost certainly be right. But I think this kind of activity is actually even more wasteful in China than in Chicago because much higher productivity levels in the US mean that the resulting \’96 expensively acquired \’96 improvements in efficiency would be more valuable in Chicago than in China. So building ultra-modern facilities may appease the pride of local officials, but it may do so at a cost far greater than its true economic benefit.

What about cases in which there is very rudimentary infrastructure that is being upgraded as part of the 2009 stimulus package? Here too I am not sure that we should be overly sanguine about the surge in infrastructure investment. China already has excellent infrastructure for such a poor country, and well before the stimulus package it was widely accepted that there had already been overbuilding, misallocated capital, and wasted investment in infrastructure. The recent surge in investment might all be for very productive purposes whose resulting increase in production will easily pay off the true, unsubsidized cost, but this is an argument that would need an awful lot of proof before I would believe it.

It is hard to imagine that a system that was already misallocating capital on a huge scale (for example by almost any reasonable standard most SOEs are value destroyers, whose viability is only assured because of input subsidies and highly subsidized borrowing costs) would suddenly, under tremendous pressure to expand investments massively and quickly \’96 and with the understanding that all risks would be socialized \’96 could do so without increasing the number of unprofitable investments. Maybe I will prove to be wrong, but I do think a lot of skepticism is warranted.

By the way my argument is not that \’93Keynesian\’94 spending is a waste. I think its usefulness depends on existing capacity use, including employment, and can generate more value for the economy than it costs. My argument \’96 a much more limited one \’96 is only that infrastructure spending is not automatically more economically viable in poor countries than in rich countries. The larger possible \’93nominal\’94 improvement in the quality of infrastructure will only lead to greater economic value if the poorer country is able to capture as much economic benefit from the investment as the richer country.

If labor productivity is much lower, as it is in China, it might not be able to do so. In fact I would go further. State-of-the-art infrastructure in China is almost always harder to justify economically than in Japan.

The coming clash in savings

September 21, 2009 in Consumption and production by Michael Pettis | No comments

Perhaps it has been because I have been so busy in meetings and school in the past week but it seems to me that not a whole lot has happened to give us much more sense of what is happening in China since the big release of economic data by the National Bureau of Statistics on Friday 11.\’a0 As I wrote then, the data was able to confirm both those who see the stimulus package has having been a big success in protecting China from the ravages of the global economic crisis as well as those who worry that it is actually making the Chinese imbalances worse.\’a0 Both are right, in my opinion.

In that light there was an interesting article in the Financial Times suggesting that the CBRC, as it has been all year, almost alone among the official institutions, is still worrying out loud and publicly about the impact of the stimulus package on the banking system:

China\’92s top banking regulator on Friday warned of growing risks to the country\’92s financial system as a result of an unprecedented expansion in new loans and urged the country\’92s lenders to improve their internal management.

The statement by Liu Mingkang, chairman of the China Banking Regulatory Commission, may signal a more assertive stance from the body in the build-up to a top-level Communist party meeting scheduled for November that will set the country\’92s economic agenda for the coming year.

The FT quotes Mr. Liu as writing that \’93This year, all kinds of risks have arisen in the banking sector along with the rapid credit expansion. Banking institutions should always stick to the bottom line of compliance management, to lay a solid foundation for risk management.\’94

\
Xinhua also carried the story in an article Friday although, perhaps not surprisingly, they seemed to play it a little softer and give it a more postive spin:
\

China’s banking regulator has reiterated that domestic lenders should seek to enhance their risk management and stick to regulatory requirements to reduce worries over financial risks caused by rapid credit growth this year.\’a0 “With bank loans growing rapidly, all kinds of risks are rising in the banking industry”, Liu Mingkang, chairman of the China Banking Regulatory Commission, was quoted as saying by Saturday’s China Daily.

Although Mr. Liu’s statements will not come as a surprise to senior policymakers, i suspect that everyone’s attention was focused on last week’s annual plenum of the Communist Party Central Committee.\’a0 Something many people were expecting to happen did not happen, leading to a whole lot of speculation about whether or not this has significant implications for a factional disagreement within the top leadership.\’a0\’a0 The always insightful Australian journalist John Garnaut had this to say:

Palace intrigue has swept Beijing following the failure of President Hu Jintao’s assumed successor to receive a crucial promotion. Vice-President Xi Jinping was expected to be promoted to Central Military Commission deputy chairman at last week’s annual plenum of the Communist Party Central Committee.

Mr Hu was given the position at the equivalent stage of his career. Mr Xi is still considered the most likely candidate to succeed Mr Hu, but his path now appears to be contingent on a period of bruising deal-making.

\
I am not smart enough to say whether Xi’s failure to get the promotion was indeed as significant as many of my Chinese and foreign friends seem to think (many also disagree), but it adds to constant rumors about factional differences, disagreements among policymakers, and other noise that is clouding our ability to understand what policymakers may do next.
\

\
I have been working on a few pieces about what i am increasingly thinking as a possible global savings clash.\’a0 To summarize briefly, we know that savings and investments must balance.\’a0 If it does not balance domestically, it balances globally through adjustments in trade and capital accounts, so that countries with excess savings (over investment) export capital to countries with excess investment.\’a0 of course to do so they must also run current account surpluses.
\

\
This, in a nutshell, is the relationship between China and the US.\’a0 In China savings had reached the highest rate, probably ever recorded, while in the US savings declined to extremely low rates.\’a0 Both were possible because China ran a large and growing trade surplus with the US.
\

\
I have always argued that high Chinese savings were a result of specific industry and trade policies that favored investment and manufacturing and that subsidized them through sluggish wage growth, low deposit rates, and other “taxes” on household income.\’a0 Obviously if production grows significantly faster than consumption, the result is a rising savings rate.
\

\
That means that for global savings and investment to balance, any change in one of the major net savers or dis-savers will have an effect on others.\’a0 As i have written many times before, I expect that one consequence of the crisis will be a deleveraging of US households and an increase in the savings rate.\’a0 This can be forestalled for a while by government dis-savings, but i don’t think government borrowing can forever hold back the necessary rise in the US savings rate.
\

\
Unless global investment rises significantly, an increase in US savings must come with a decrease in non-US savings, and in practice this means Chinese savings.\’a0 If global investment declines, of course, this is even more true.\’a0 But I do not believe that it will be easy for the Chinese savings rate to decline, for all the reasons i have mentioned many times before, especially once the impact of the stimulus package wears off.
\

\
So how does this clash over global savings get resolved?\’a0 Obviously there are lots of ways it can get fixed.\’a0 One way, of course, is a completely benign surge in Chinese consumption.\’a0 Another way is a slowdown in global growth.\’a0 The first is unlikely.\’a0 The second almost by definition means a huge increase in trade tensions since it is through the trade account that the global slowdown will be distributed.\’a0 Needless to say every country will be eager to pass on the adjustment to other countries, and in this kind of fight I am afraid trade surplus countries are more vulnerable than trade deficit countries.
\

\
On a related note, several investors i have met in the past two weeks have been very worried about some seemingly bizarre reports that apparently argue that the dollar is in for a very difficult period over the next few years because of the Chinese adjustment.\’a0 As far as I understand it, the claim is being made that the Chinese trade surplus is destined to fall rapidly over the next few years.\’a0 So far so good.\’a0 I agree.
\

\
The report then argues that because of the fall in the Chinese trade surplus, the PBoC and other Chinese institutions will be buying fewer dollar assets.\’a0 Again, I agree.\’a0 This is simply the statement of an accounting identity.\’a0 But here is the bizarre part.\’a0 The report then claims that such a massive decline in dollar purchases by Chinese investors is very bearish for the dollar.\’a0 If one of the world’s biggest buyers of dollars stops buying, in other words, the dollar must decline.
\

\
On the face of it this might seem true, and even obvious, but a very quick glance at the balance of payments suggests that this view seriously confuses the nature of dollar flows.\’a0 For many years a lot of commentators argued that the dollar was destined to collapse because of the large US trade deficit, which had the effect of forcing a huge flow of dollars onto foreign investors.\’a0 Since the flow of dollars was likely to exceed the appetite for dollars, this would unquestionably put downward pressure on the dollar.
\

\
In the standard trade model this would almost certainly be true, but Jim Walker, of Asianomics, manged to undermine the argument with his “51st state” thesis.\’a0 He argued that this would only be true if the US ran a current account surplus with non-dollar bloc countries who weren’t forced to recycle their surpluses because of the nature of their currency regimes.\’a0 If they didn’t recycle, in other words, the dollar would weaken and the current account surpluses and deficits would disappear.\’a0 As long as they were forced to recycle, however, this wouldn’t occur.\’a0 If you combine all the dollar bloc countries (China, much of Latin America, etc.) into a “51st state”, Walker concluded, the US in fact ran no current account deficit and so there was no balance of payments pressure on the dollar.
\

\
I think Jim walker is right, but I also agree that when a country is running large current account deficits it is natural to worry about a depreciation in its currency as the resulting outflow through the current account leaves investors less willing to continue holding more.\’a0 That is why the new argument for dollar appreciation is, in my opinion, a little weird.\’a0 It is true that China will soon be buying a lot fewer dollars, but this will happen presumably because a rapidly declining US trade deficit is forcing the decline in the Chinese trade surplus.\’a0 That means that although the demand for dollars by foreigners will decline, it will decline at the same time and for the same reason that the supply of dollars also declines.\’a0 there isn’t likely to be any imbalance between supply and demand.
\

\
I have no idea which way the dollar is going to go over the next few years, but I am pretty sure that if it declines it will not be because the Chinese are buying fewer dollars as they recycle their smaller trade surplus.\’a0 That doesn’t make sense to me.

China\’92s August data confirms both optimists and pessimists

September 11, 2009 in Consumption and production, Fiscal stimulus by Michael Pettis | No comments

Regular readers of my blog will have noted all sorts of unfortunate goings on here in recent days. \’a0It has become impossible to get into the comments section, or indeed into any other section of my blog except the front page, and so to my great dismay the excellent discussions that have been so useful for me have been temporarily halted. \’a0I am not sure why this is the case, and now that school has started again I hope to get one of my terribly smart Peking University students to find out and fix it. \’a0

\’a0

Actually in recent months a large number of China scholars who I know \’96 and me too \’96 seem to have been targeted by very specific viruses, with emails and attachments cleverly disguised to look like something we would want to read from someone we would trust.\’a0 I am not smart enough to know what those viruses do, but I have been warned that they probably allow someone access into my computers. \’a0

\’a0

Perhaps for the same reason my own blog has been hacked in some way, but it seems to me that if anyone really wanted to close me down they would have closed down the blog altogether and not just the comments section, so unless there is a commenter that regularly raises the ire of some censor out there, I suspect the problem has more to do with my blog site than with any malicious intent.\’a0 Of course if my blog suddenly begins posting pornographic pictures, spewing venom, or otherwise does some unexpected and obnoxious things, please know that it was probably not me who came up with the idea.

\’a0

I have also had a lot of trouble getting my regular proxies that allow me to jump the great firewall and post onto my blog. \’a0That is why my posting has slowed a bit, but it seems that yesterday and today the anti-proxy regiments have been at least temporarily defeated.

\’a0

That\’92s good, because of course today the National Bureau of Statistics has released a whole lot of data.\’a0 A European TV station asked me to comment on the import of the data, and while I hate to make too much of a few data points, I was able to say rather glibly that the data pretty much confirmed the hopes of the optimists as well as the fears of the pessimists. \’a0

\’a0

I will explain why I think this below, but I should note that most analysts were pleased with the results, and the stock market surged on the news.\’a0 The SSE Composite was up 2.2% today.\’a0 Surprisingly it was down 0.7% yesterday, suggesting that there was probably no information leakage.\’a0 Things seem to be improving on that front.

\’a0

To summarize the data released today, manufacturing output was up by 12.3% year on year, better than last month\’92s 10.8% and higher than consensus, although I think last August, during the Olympics, a lot of factories were closed so that this number may not be as impressive as it seems.\’a0 Steel output was up 29% and auto production was up 90%, which as my friend Mark Williams at Capital Economics points out is not likely to soothe worries about overcapacity creation.

\’a0

Urban fixed-asset investment was up 33.0% for the first eight months of the year, which slightly exceeded already-high estimates of 32.5%.\’a0 This suggests that it is still investment that is in the driver\’92s seat, as far as growth is concerned.\’a0

\’a0

This might not be as obvious as all that.\’a0 A lot of people were excited that retail sales climbed 15.4%, slightly higher than consensus and the highest growth rate all year after seasonal factors are stripped out, but remember that retail sales are not a very good proxy for consumption growth. \’a0Also remember that this surge in liquidity can easily cause consumption to rise in a temporary way without indicating anything structural about changing consumption and saving patterns in China. \’a0In 1988-89 consumption in Japan also surged, probably as a consequence of the investment boom, but it was unable to survive, if I remember correctly, the contraction in that boom in the 1990s.

\’a0

The most interesting piece of information is that net new lending for the month was RMB 410 billion, less than half the monthly average this year (RMB 1,105 billion).\’a0 This seems small given the huge numbers we\’92ve seen but, as I pointed out two weeks ago, last August new lending was around RMB 272 billion, and if you strip out the bills coming due the real increase in medium- and long-term lending is closer to RMB 550-600 billion. \’a0More importantly, RMB 410 billion is a lot more than the rumors of RMB 300 billion that had panicked the market last week.\’a0

\’a0

The one piece of news that everyone read as negative was the trade data. \’a0Exports were down 23.4% and imports down 17.0%, both substantially worse than expected, although leaving the trade surplus at a still-hefty $15.7 billion, which is roughly average for the year.\’a0

\’a0

For the optimists, the economic numbers, with the exception of the trade data, were all positive and suggest that China is on track to recovery.\’a0 For them, the great risk to China was that the global contraction in demand would result in terrible damage to China\’92s export industry and, with it, would cause factory closings and soaring unemployment.\’a0 Rising unemployment would lead to a collapse in consumption, and of course would not make China\’92s transition easier

\’a0

The main purpose of the stimulus package, in this view, was to forestall an economic contraction and with it the possibility that the economy would fall into an ugly process in which rising unemployment would cause a contraction in Chinese consumption which, when added to the contraction in foreign demand for Chinese exports, would push the economy into a tailspin.\’a0 In that sense the stimulus has proven to be a great success.\’a0 Chinese growth has slowed, but by a lot less than expected, and unemployment seems to be manageable.\’a0 The August data points pretty solidly to continued growth.

\’a0

And yet, and yet\’85.

\’a0

For pessimists like me the global contraction underscored Chinese vulnerability to out-of-control US consumption, and the need to develop a more balanced approach in which Chinese consumers take a larger share of China\’92s production.\’a0 This vulnerability existed in large part because China was overly reliant on investment for its growth.\’a0 China has had probably the highest investment rate ever recorded for a large economy, and for years there has been widespread concern that much of this investment was misallocated.\’a0

\’a0

It is only because the cost of capital is artificially so low (thanks to Chinese households, who are forced to earn a miniscule return on their savings) that many companies are able to show profits at all. \’a0A few months ago I wrote about an HKMA paper that suggested that the implicit interest-rate subsidy to SOEs \’96 not relative to the \’93right\’94 interest rate in China (whatever that may be but which is certainly many percentage points higher than the official lending rates) but relative to the borrowing cost of large Chinese private corporations \’96 accounted for 100% of SOE profitability. \’a0If China had reasonable interest rates, in other words, (and in fact there were negative real rates for much of the recent past), SOEs would on average be value destroyers.

\’a0

This, by the way, is why China\’92s supposedly puzzling addiction to capital-intensive growth rather than labor-intensive growth \’96 more befitting to an economy with lots of unskilled labor and very poor technology \’96 is not so puzzling. \’a0If you artificially lower the price of a particular input, it is not surprising that producers will use more of that input than might otherwise be considered optimal.\’a0 With capital practically free, capital is everyone\’92s favorite input in spite of incredibly low labor costs.

\’a0

With the recent surge in government financed investment (and I include most bank lending in this category), it would be surprising to me if much of this year\’92s new investment were not of even lower quality than the older investment, with very low or even negative expected returns.\’a0 If this turns out to be true, it means that the only way these investments could be viable is by effectively continuing to \’93tax\’94 Chinese households to subsidize state-owned enterprises and large manufacturers.\’a0 This tax of course will come mainly in the form of low wage growth and extremely low deposit rates on the savings of Chinese households.

\’a0

This is why we all hope Chinese growth will become more reliant on rising consumption rather than on rising investment, much of which is certain to be unprofitable.\’a0 The current path requires a large trade surplus to absorb the difference between what China consumes and what it produces, but it is not clear that foreign consumers will absorb the balance.\’a0 China is trying to plug the gap by a surge in government-financed investment, but this is likely only to widen the gap in the future.\’a0

\’a0

So the August data suggests that while China is growing, it is actually more reliant, not less reliant, on investment.\’a0 What is worse the very poor import numbers suggest that in spite of high retail growth figures, consumption growth in China is still sluggish.

\’a0

For the pessimists, then, the August numbers confirm that the stimulus package may be boosting production solely because of government-financed investment, and that a serious misallocation problem will result in more future pressure on Chinese households to foot the bill.\’a0 The export numbers show that China\’92s external accounts continue to deteriorate, and it will take more than simply an end to the global crisis to return to the good old days.

\’a0

So who is right, the optimists or the pessimists?\’a0

\’a0

In fact both are right.\’a0 If the purpose of the stimulus package was solely to protect China from the immediate employment impact of the global contraction in demand, it has been an almost unqualified success.

\’a0

But if at least part of the goal was to help China shift its unbalanced growth model to one less reliant on foreign, and especially American, consumers, it is not clear that any progress has been made.\’a0 In fact to the extent that a significant share of new investment has been wasted, it may actually make future imbalances worse.

\’a0

China\’92s response to the global crisis needs to be seen as a two-part process.\’a0 The first part is to goose economic growth in response to the rapid deterioration in the external environment.\’a0 The second part is to rebalance the economy away from its excess reliance on investment and foreign demand.\’a0 The August data seem to confirm that China is very successfully managing the first part.\’a0 Whether it has made any progress on the second part is still very much open to question.

\’a0

It\’92s not yet the end of China\’92s massive stimulus

August 31, 2009 in Consumption and production, Fiscal stimulus by Michael Pettis | 31 comments

\’a0According to a recent article on Reuters, on Saturday Lou Jiwei, the chairman of the CIC, China\’92s sovereign wealth fund, said at a conference on Saturday in response to a question about his expected performance: \’93It will not be too bad this year. Both China and America are addressing bubbles by creating more bubbles and we’re just taking advantage of that. So we can’t lose.\’94
\

\
In my last entry I noted that after the recent \’93green shoots\’94 period, during time which it seemed hard to find anyone who was skeptical of our seeming ability to turn the corner on the crisis without actually having addressed any of the underlying imbalances, it was good to see that more and more analysts, and especially policymakers, had begun to worry again.\’a0 President Hoover went down in a blaze with his \’93light at the end of the tunnel\’94, and of course one of my favorite stories of that time is his response in June 1930 to a delegation requesting a public works program to help speed the recovery: \’93Gentleman, you have come sixty days too late. The depression is over.\’94
\

\

\

As I see it the more policymakers worry, the better. This crisis is far from over. Until we know how the continued adjustment in US household consumption and debt will evolve, and how this adjustment will play out in China\’92s own changing consumption rate \’96 most importantly whether it will complement the fiscal and credit expansion embarked upon by Beijing or, as I believe, conflict enormously with it \’96 the crisis won\’92t be over. We need policymakers to resist the green-shoots nonsense and to worry about what happens when fiscal, monetary and credit tools stop working.

Although I thoroughly disagree with the \’93So we can\’92t lose\’94 part of Mr. Lou\’92s statement \’96 I have been a trader for too long to hear those words with anything but the deepest dread, and I am sure he didn’t intend\’a0the way it read\’a0\’96 it is nonetheless interesting to me that by now skepticism is so widespread that a major investor can even propose our inability to work through the imbalances as a reasonable investment strategy.

We need skepticism. For one thing it has caused Beijing increasing worry about the risks of continuing to extend the stimulus package, to the point where they are now making serious noises about cutting back. My biweekly column in today\’92s South China Morning Post argues that in spite of the damage this has done to the stock market, it is undoubtedly a good thing that they are thinking about cutting back.

So Chinese policymakers have had to choose between policies that boost employment in the short term while making the overcapacity problem in the long term worse and, on the other hand, force a more efficient adjustment in the domestic imbalance while increasing job losses.

\’a0

Until now, Beijing had come down resolutely on the side of boosting employment. It had shifted a massive amount of resources, mainly through the banking system, into new investment in infrastructure and new production facilities. This created jobs and boosted consumption, but it did so by expanding current and future production even faster, only worsening the domestic imbalances and making China even more reliant on US consumption.

\

\

It probably had no choice. As in nearly every major economy, the first instinct of policymakers since the crisis began has been to enact measures to slow unemployment growth. If unemployment grew too quickly and caused consumption to fall, it could easily tip the economy into a long-term and irreversible contraction.

\’a0

But there was always a limit to how far Beijing should push. It could continue spending like crazy on good and bad projects to keep workers employed, but if all this spending simply increases capacity faster than it raised consumption, the net result would be an unsustainable debt burden and a more difficult reckoning.

\

\

That is why we should welcome the signs that Beijing may be reaching the limits of its investment push. The government believes that it has created enough momentum to avoid the worst consequences of the global crisis and the contraction in the export markets, but it is also stepping back from creating a worse crisis.

But it won\’92t be easy, and I suspect that already the effect of rumors about slowing the fiscal expansion is strengthening the hands of those who want to stomp again on the gas pedal. For example the stock market was down 6.7% today, bringing its total decline since August 4 to 23.3%. Even my superstar PKU student Gao Ming, who has so far ridden this chaos pretty well, admitted to me today that it was not a good day for him.

Why did the market collapse? Forget about fundamentals. As I have argued many times before, China lacks the necessary tools that fundamental investors use (e.g. good macro data, good financial statements, a clear corporate governance framework, a stable regulatory environment, a market discount rate) and so no matter what people say, there are no fundamental investing here. There is only speculation, and the two things above all that drive the markets are those old speculator favorites, changes in underlying liquidity and government signaling.

The whole market is worried about both, and the most important is concern that the days of explosive bank credit growth are behind us. On Friday, for example, Bloomberg reported that:

Bank of China Ltd., the nation\’92s third-largest by assets, plans to slow credit growth in the second half of the year and improve loan quality after posting an unexpected profit gain in the second quarter.

\’a0

\’85Lending in the second half will be \’93much smaller,\’94 with new credit in July and August dropping from the monthly averages of the first half, President Li Lihui told reporters yesterday.

Today the mainland newspapers were even more worrying. Several reported that new loans in August would be just RMB 300 billion, after last months\’92 new loan total of RMB 356 billion, and RMB 1,231 billion on average during the previous six months.

RMB 300 billion is nothing to sneeze at, especially since that probably nets out a lot of bills coming due \’96 so that new medium-and long-term investment is likely to be substantially higher. It is also worth remembering that August is normally a bad month for new lending \’96 last year net new loans were only RMB 272 billion.

Still, after the deluge of new lending for the first half of the year, it clearly represents a significant contraction in the rate of credit expansion, and if you believe, as I do, that China\’92s \’93impressive\’94 growth rate this year is actually a very disappointing consequence of a huge fiscal and credit stimulus, any indication that the stimulus will slow down cannot be good for sentiment.

I wonder, and I know I am not the only one wondering, what Zhongnanhai is thinking as it sees the impact of these rumors of a contraction in the furious rate of credit expansion. For one thing it seems that there are only two positions on the switch \’96 \’93surge\’94 and \’93swoon\’94 \’96 and I suspect that very quickly we will see the switch turned back to \’93surge\’94. Although there seems to have been a little upward blip in US import numbers, I think this represents more of a temporary bounce from a steep earlier decline, and that the external environment continues to be very poor.

My guess is that if the local stock markets do not soon recover their bounce (and they won\’92t without government help) and, even worse, if we start to see the awful sentiment seep into the real estate sector, Beijing will once again push forcefully for credit and fiscal expansion. In my opinion there is simply no way that domestic consumption \’96 unless it is primed with government giveaways \’96 can make up the slack quickly enough.

Speaking of which I saw an interesting article in today\’92s People\’92s Daily. On the one hand it seems positive for an eventual generational-inspired rise in consumption, and on the other hand it seems negative about structural impediments:

College students, once a major demographic for banks issuing credit cards in China, are now finding that many lenders such as China Merchants Bank and Bank of Communications have recently steepened their application requirements or stopped issuing credit cards to students altogether.

\’a0

The changes in policy originate with a notice issued by the China Banking Regulatory Commission at the end of July. According to the notice, other than parents authorizing access their account, banks are not allowed to issue credit cards to those under 18. For students over 18 unemployed or without income, a cosigner is required.\’a0\’a0 Paying with plastic is really common on campuses, and is not unusual for a student in China to have up to 3 to 4 credit cards.\’a0\’a0“Whenever I go back home, I use a credit card to buy plane tickets, because at the end of the semester I’m usually short on cash,” said Sun Chenghao, a senior student at the China Foreign Affairs University.\’a0\’a0

\
\’a0
\

But such convenience also has its drawbacks. Of all recent credit card debt cases heard at the People’s Court in Beijing’s Xuanwu District this July, about 25 percent involved college students.

\’a0

\

\

\’a0

\
\’a0
\

\

\

\’a0

\

\

\’a0

\
\’a0
\

Yet another discussion on the Asian savings glut hypothesis, and why it matters

August 20, 2009 in Asian development model, Consumption and production, Exports and imports, Policy, Savings glut, Stock market by Michael Pettis | 84 comments

The Shanghai stock market was up 4.5% in very nervous trading today but down 16.3% since its recent peak at 3478 on August 4, and still trading at more than 30 times earnings.\’a0 All this turmoil is triggering all sorts of worried comments about the sustainability of the fiscal stimulus package and whether it has already reached the end of its effectiveness (it hasn\’92t \’96 the government still has credit and fiscal firepower, and will use it if growth slows down significantly in the next few quarters).\’a0 It also makes it harder, but probably more useful than ever, to focus on the bigger picture, and this entry is definitely big picture.\’a0 It also turned out to be a very long piece, as these big-picture pieces tend to.\’a0

The topic is whether or not the global imbalances that have led to the current crisis were in any way \’93caused\’94 by the Asian savings glut, and besides arguing why I think this may be the case, I want also to argue that getting this argument right is far more important than many seem to realize. \’a0Mohamed Ariff, executive director of the Malaysian Institute of Economic Research, indirectly suggests why in a good OpEd article in today\’92s New Strait Times:\’a0

China\’a0is seen as the beacon of hope in these days of gloom and doom. This has led some observers to think China will lead East Asian economic recovery and thereby spearhead a global economic turn-around. But this faith in China as saviour may be misplaced.China’s imports from the rest of East Asia consist mostly of raw materials, intermediate products and components and parts, the bulk of it turned into manufactures for exports. China’s imports of consumer products from the region account for no more than a small proportion.

\

\

China’s imports from its neighbours have plummeted in the wake of the slump in China’s own exports, although the Chinese economy is growing at seven to eight per cent, because China depends largely on domestic production for its own consumption, which does not spill over to the rest of the region through trade.

\

\

Therefore, China will import more only if it can export more. For this to happen, the demand for China’s exports in the US and European markets must first recover.

\
\’a0
\

\

\

But our definition of a \’93recovery\’94 in the US, and whether it will indeed happen in the way that Ariff requires for Asian growth to return, depends in an important way on whether or not the current imbalances were caused primarily by an original distortion in US consumption or in Asian savings.\’a0

I started writing this because while googling around looking for something else, I stumbled early this week upon a blog by LSE\’92s Danny Quah with the intriguing title \’93Where in the world is Asian Thrift and the Global Savings Glut?\’94\’a0 I later found that like mine, his blog is carried by Nouriel Roubini\’92s RGE Monitor.\’a0 I also subsequently discovered by a weird coincidence that on Saturday I am sharing a panel with him in a conference at the Guanghua School at Peking University, where we will be discussing the \’93Reconstruction of Global Finance\’94.\’a0\’a0\’a0

The whole \’93savings glut\’94 debate is a controversial one because almost from the start it has degenerated into a fairly silly argument about who to blame for the global imbalances and the subsequent crisis \’96 or more specifically and more excitingly, whether the predator was wholly the foolish American consumer or the beetling Chinese saver.\’a0 Three months ago Brad Setser discussed all this in one of his blog entries that (inevitably) drew more comments than most, and as usual he provides a concise and enlightening discussion on the subject which you might want to read.\’a0 He is a proponent of the hypothesis, but nonetheless pretty fair-minded.\’a0

Professor Quah weighs in on the other side of the savings glut debate although, unlike most others in the debate, he seems not terribly concerned about assigning full blame to any of the major parties.\’a0 It is neither excess US consumption nor excess US savings that solely \’93caused\’94 the imbalance, in other words, because necessarily both sides are required for it to exist.\’a0

Except for the possibility of trade with outer space, the US deficit has to be matched dollar-for-dollar by trade surpluses in the rest of the world. Correspondingly, therefore, the rest of the world has been saving\’97consuming less than it has been producing\’97and accumulating dollar claims against the US as a result.\’a0

In this description, however large the global imbalance, a savings glut\’97wherever or however it might arise on Earth\’97has no independent existence. It makes as much sense to say the world\’92s excess savings caused enthusiastic US consumers to flood into Walmart to buy $12 DVD players, as to say US consumer profligacy made hungry Chinese peasants abstain even more and instead plow their incomes into holdings of US Treasury bills.\’a0\’a0

When two variables have always-identical magnitudes, obviously neither can usefully be said to cause the other.\’a0\’a0

Who are the predators?

This is correct, but as an aside, the discussion about enthusiastic American consumers forcing the Chinese to save, or hungry Chinese savers forcing Americans to consume, typically uses colorful but totally inappropriate images to describe the dynamics of the this process.\’a0 For example, I often hear opponents of the Asian savings glut hypothesis say, voices dripping with disbelief, that the savings glut hypothesis insists that the poor American consumer rushed out to buy another DVD player because terrible China forced him to borrow the money and buy the DVD player.\’a0 How could that possibly happen?\’a0

Well, that\’92s not how it would have happened.\’a0 In any large country, there are millions of households able and interested in increasing savings or in increasing borrowing.\’a0 Specific policies or financial conditions will determine at any given time changes in the behavior of some of these individual households, so that at the macro level, and only at the macro level, the country will have seen an increase in savings or an increase in consumption.\’a0\’a0\’a0

It is not every household that rushes out to consume when consumption rises, and this never happen because predatory savers force an otherwise unwilling consumer to buy. \’a0So if it had indeed been rising Asian savings that drove the US consumption binge, policies aimed at constraining Asian consumption and boosting Asian production (which cause savings to rise) will have initially led to a rising Asian trade surplus and US trade deficit, as the tradable goods sector in Asia expands and the tradable goods sector in the US contracts.\’a0

This surplus would be recycled into the US via purchases of highly liquid securities.\’a0 If the Fed failed to respond to this increase in liquidity by raising interest rates and contracting money (and contracting the tradable good sector), the financial system would have to accommodate the rising liquidity as it has always done throughout history \’96 by growing financial balance sheets and taking on more risk.\’a0 In that case the conditions for consumer borrowing will have been made increasingly easy, and those households who needed or were predisposed to borrow under easier lending conditions, and pressure on the parts of banks to extend credit, will do so.\’a0\’a0\’a0

As long as there are some households willing, however appropriately or foolishly, to increase consumption, the easier availability of consumer credit will cause them to increase consumption \’96 this has happened many times and in many countries, and has nothing to do with a predisposition to excess consumption.\’a0 Furthermore as recycled liquidity boosts household wealth by boosting the value of homes and investment portfolios, the rising wealth of each individual household will have an impact similar to rising income \’96 and with it consumption will rise.\’a0\’a0\’a0

So the point is the not very controversial suggestion that a surge in domestic liquidity in the US can easily cause US consumption to rise.\’a0 If that liquidity surge was \’93caused\’94 by the recycling of a large and growing trade deficit, then it is easy to see how at the macro level US consumption would rise in response to a surge in Asian savings.\’a0

Similarly, the proponents of the Asian savings glut hypothesis wonder in disbelief how an American consumer deciding to buy a DVD player could have possibly \’93forced\’94 poor Chinese peasants to cut their already minimal consumption and increase their savings. \’a0But there was no force.\’a0 A sudden explosion in binge consumption in the US would divert production from China, and as China increased the share of its output dedicated to exports, total production would not immediately be matched by total domestic consumption (Americans bought some of it) and the Chinese savings rate would necessarily increase \’96 whether at the household level or at the corporate or government level.\’a0

The interest rate argument\’a0

The point is that sarcastic comments about predatory American consumers forcing dim-witted Chinese households to save more and consume less, or predatory Chinese savers forcing helpless American households to borrow and consume, may be good debating tactics but they are misleading and explain nothing.\’a0 At the macro level either event \’96 higher Asian savings leading to higher American consumption, or higher American consumption leading to higher Asian savings, or even a combination of the two \’96 is perfectly possible.\’a0

So why should we accept the Asian savings glut hypothesis?\’a0 One argument that I first saw proposed by Brad Setser was that if the imbalances had been driven by US consumption, and therefore US borrowing needs, the consequence should have been an increase in US interest rates.\’a0 Had they been driven by excess savings, US borrowing rates would have probably declined.\’a0\’a0

In fact during most of the relevant period US interest rates did decline, even leading to the US Fed several times complaining about its inability to control domestic long-term rates.\’a0 So that pretty much settles it, right?\’a0 But Professor Quah dismisses this argument:\’a0

Many other factors could, of course, have driven down short rates: US monetary policy responded to national economic downturns in 1991 and 2001. Through the 1990s inflation rates worldwide converged and fell, together with short-term interest rates set by central banks everywhere. The burst of the dot-com bubble in March 2000 saw the NASDAQ index decline 77% in the following 18 months, prompting action by the US Federal Reserve. Japan\’92s monetary policy during its decade-long recession drove nominal interest rates there to zero.\’a0

Although he is right, this is not a completely satisfying dismissal.\’a0 The same savings glut that pushed down US interest rates could easily have pushed down global interest rates, especially in a world that was seeing rapidly rising capital flows that in many cases were aimed at \’93arbitraging\’94 (absolutely the wrong word, of course, but one widely used in the markets at the time) interest rate differentials.\’a0 After all it is often the case that, especially during periods of large international movements of capital, increases or reductions in US interest rates (or in British rates during the globalization period at the end of the 19th Century) are matched by changes in foreign interest rates.\’a0

Still, the fact is that his response does show that the interest rate argument is not final.\’a0 There might be other perfectly good reasons that explain the decline in US interest rates.\’a0

The bilateral trade argument\’a0

Quah\’92s main argument against the savings glut hypothesis, at least as far as his blog entry, seems to be that it could not have been a rise in Asian savings that drove the global imbalances because had it done so, much of the imbalance would have rested between Asia (or China, more specifically) and the US.\’a0 The strongest piece of evidence he presents is a chart that shows the US bilateral trade balances between the US on one side and China, developing Asia, the EU, and oil exporters on the other.\’a0 I have reproduced the graph below, but if you can\’92t see it well, just click on Quah\’92s blog (blocked in China, so China-based readers will need to use a proxy), and click on the graph itself for an enlargement (I wish I was clever enough to do things like that).

\

\

\’a0

As the chart shows, the US trade deficit rose nearly as quickly, or even more so, with those other regions as it did with China and/or developing Asia.\’a0 It wasn\’92t just a US-China phenomenon or a US-Asia phenomenon, it was a US-everybody phenomenon.\’a0

Quah\’92s argument seemed to be a powerful one at first, and I had to think about it for a while or else I would have to find myself deserting from the \’93savings glut\’94 camp.\’a0 In the end, however, I think his argument it turns out not to be very satisfying and I still think it runs against a timing story that better explains the imbalances.\’a0 I\’92ll say more on that later, but it seems to me that in a \’93globalized\’94 world, if the Asian savings glut hypothesis is true, not only would rising bilateral trade deficit between the US and other countries outside of developing Asia be possible, but they would even be almost necessary.\’a0

Why?\’a0 Because we have to be careful about misreading bilateral trade numbers.\’a0 It is the aggregates that usually matter.\’a0 I don\’92t have the data in front of me, but I believe that Europe did not run significant and rapidly growing aggregate trade surpluses during this period.\’a0 If that\’92s the case, then a growing bilateral surplus with the US is perfectly consistent with the savings glut hypothesis as long as you assume that trade is international and that any specific product can be produced and assembled in many countries \’96 which is of course a pretty unremarkable assumption.\’a0

So, for example, if rising Asian net savings \’93caused\’94 rising American net consumption (in the way described above \’96 no sarcasm, please), it would mean that money recycled from Asia into the US caused the US trade deficit to rise as it was intermediated by the financial system into consumer financing, even as it caused Asian trade surpluses to rise.\’a0

It\’92s the aggregate balance that matters\’a0

But, and this is the important point, the trade did not need to occur only at the bilateral level.\’a0 If rising Chinese savings was intermediated into rising US consumption and this bilateral relationship was resolved, to take a concrete example, by Chinese exporters producing shoes and American consumers buying shoes, the trade would not have had to occur directly between the two.\’a0 When Americans shop for shoes, they don\’92t care which country saw net savings rise, and when Chinese sell shoes they don\’92t care whose economy saw an increase in net consumption.\’a0 China could have produced shoes, sold them to a designer in Italy, where they would be packaged and branded, and then sold to American consumers.\’a0

In this simple case, Chinese excess savings would have \’93caused\’94 Americans to borrow money and buy the shoes, and so China would run a trade surplus, the US would run a trade deficit, and Italy would be balanced.\’a0 But Italy would nonetheless show a bilateral surplus with the US and a bilateral deficit with China.\’a0\’a0\’a0

Excess US consumption, in other words, would still have been \’93caused\’94 by excess Chinese savings in this case, but global trading and processing networks would have the bilateral trade imbalances, and their countervailing obverses, spread out though the world.\’a0 Many countries would run surpluses with the US and deficits with Asia, but at the aggregate level they would balance out at close to zero, and the US would be left with the sum of its bilateral deficits and Asia with the sum of its bilateral surpluses.\’a0

The point is that there is nothing in the Asian savings glut hypothesis that requires that all trade imbalances occur at a bilateral level and only between the participating countries \’96 that the deficit/surplus imbalances occur between the US and Asia.\’a0 It only requires that the US, as the equilibrator to rising Asian savings, have a large and growing trade deficit and Asia have a large and growing trade surplus.\’a0 If other regions also have large and growing aggregate trade surpluses that fed into the US deficit at the same time, that would perhaps be the problem Quah says it is, and either would need to be explained or would create problems for the hypothesis.\’a0 But they didn’t.\’a0

With one big exception, of course.\’a0 Oil exporters did see not only rising bilateral trade surpluses with the US, but they also saw rising aggregate surpluses.\’a0 Does this somehow weaken the savings glut hypothesis?\’a0 Again no, because those surpluses reflect one thing only, rising oil prices, and in an environment of rapid US and Asian growth, we would expect oil (and other commodity) prices to rise.\’a0 In fact the savings glut hypothesis would predict that as long as the recycling was occurring efficiently, both countries would grow quickly and high commodity prices would be not only possible, but in fact highly likely.\’a0

So as I see it, this is how the arguments and counterarguments stand:\’a0

1.\’a0\’a0\’a0\’a0\’a0\’a0\’a0 The argument that declining US interest rates proves the correctness of the savings glut hypothesis is wrong.\’a0 Declining US interest rates are suggestive but not final.\’a0 Other things could have explained declining US interest rates during this period, and of course there is easily a possibility of feedback loops in which any initial decline in US interest rates could, by increasing household wealth (via rising asset values) increase consumption and the US trade deficit, leading to Asian recycling, and so on to more lower interest rates.\’a0

2.\’a0\’a0\’a0\’a0\’a0\’a0\’a0 The argument that rising US bilateral deficits with many regions around the world disprove that the savings glut hypothesis is also wrong, and much less suggestive.\’a0 On the contrary, if the hypothesis is correct and if trading is truly globalized, we would expect US bilateral deficits to be high with nearly everybody.\’a0 At the aggregate level, however, we would not expect anyone except the high-saving Asian saving countries to run large trade surpluses.\’a0

3.\’a0\’a0\’a0\’a0\’a0\’a0\’a0 There was also an argument that I associate with Morgan Stanley\’92s Stephen Roach \’96 a very smart man who by the way disagrees strongly with the hypothesis \’96 since he was the one who first made this argument to me, over a lunch at Peking University two years ago.\’a0 According to Roach there has been no significant increase in global savings during the savings-glut-hypothesis period, which pretty much demolishes the idea of a saving glut.\’a0\’a0\’a0

I disagree because the hypothesis doesn\’92t imply in any way that global savings have increased.\’a0 In a closed economic system, unless investment has increased commensurately, an increase in savings in one part of the system must necessarily come with a reduction in savings elsewhere, and this was exactly the point of ascribing the current trade imbalances to a forced rise in Asian savings.\’a0 Rising Asian savings \’93forced\’94 declining US savings by causing the US financial system to accommodate growing domestic liquidity by taking on risk (again, no sarcasm please \’96 you might disagree but in itself this is not implausible).\’a0

Timing the flows\’a0

So where does that leave us?\’a0 Before answering, I think there is another thing to think about here, as I wrote earlier in this entry, and that is the timing issue.\’a0

In a June 4, 2008 entry, much of which is reproduced here, I mentioned a very interesting paper by German economist Jorg Bibow of the Levy Economics Institute of Bard College (The International Monetary (Non-)Order and the \’93Global Capital Flows Paradox\’94).\’a0 In it the author considers the \’93paradox\’94 of high and rising capital flows from developing to developed countries during the past decade.\’a0 This is a paradox because most economic theory (and history) suggests that developing countries are net recipients of investment, not net providers.\’a0

Bibow rejects the Asian savings glut hypothesis, but my understanding of his paper is that he agrees with much of what I understand the theory to be but rejects it on much narrower technical grounds \’96 he claims that the saving glut hypothesis is based on the \’93fatally flawed\’94 (his words) loanable funds theory.\’a0 However his narrative (to be horribly post-modern for a moment) of events seems very close to my own.\’a0

What interests me most is the data he provides in his paper (and you can see the accompanying graphs by following the link to his paper).\’a0 First off, Bibow discusses the evolution of the US current account deficit over the past fifty years.\’a0\’a0

Basically, according to the data quoted in Bibow\’92s paper, the US current account has been within a range of a surplus of 1% of GDP and a deficit of 1% of GDP for most of last fifty years with two exceptions.\’a0 The first exception occurred in the mid-1980s when the US current account deficit rose to nearly 3.5% of GDP in 1986-87 before declining sharply and running into a small surplus in 1990.\’a0 The second exception began technically in 1994, around the time of the Mexican crisis, when the US current account deficit climbed to around 1.6% of GDP, before it began to decline again, but it really took off in 1997-98, when it raced forward to peak at around 7% of GDP in 2007.\’a0\’a0

As an aside I should add that there was an acceleration of the growth rate of the deficit around 2004, if I remember, and I have a pretty strong suspicion that this had something to do with the financing of the Iraq war.\’a0 As I have pointed out before, US asset markets and consumption often boom during unpopular wars, like the Vietnam War, which tend to be financed not with taxes but with money creation and debt, and often these two things lead to great markets \’96 for a while.\’a0

If the US trade deficit was driven simply by an out-of-control US consumption binge, it is a little hard to see why it would have followed a pattern of general stability marked by two surges \’96 a small one from 1984-88 and a very large one after 1997.\’a0 If it was driven by Asian savings, this pattern becomes a little easier to understand \’96 or at least, what amounts to the same thing, we can posit a more plausible story to explain it.\’a0\’a0\’a0

The narrative\’a0\’a0

I will ignore the 1980s surge because this post is already too long, but again one can tell a very plausible story based on Japanese trade policies and domestic savings.\’a0 The post-1997 surge is much larger and more interesting.\’a0 1997 was, of course, the year in which several Asian countries, after years of tremendous growth and what seemed like invulnerable balance sheets, experienced terrifying financial crises and viciously sharp economic slowdowns, which profoundly impressed Asian policy-makers and has affected policy decisions to this day.\’a0\’a0

Since the main cause of the crisis seemed to be the sudden reversal in the early 1990s of current account surpluses into substantial deficits, along with highly unstable balance sheets in which large external obligations were mismatched with domestic assets and \’93hedged\’94 with extremely low levels of foreign reserves, one of the main (if mistaken) lessons policy-makers learned was the need to run current account surpluses and to amass large foreign currency reserves to protect countries from a repeat of the disastrous crisis of 1997.\’a0

These countries, consequently, but into place \’93mercantilist\’94 policies in order to achieve both goals \’96 persistent trade surpluses and large amounts of foreign currency reserves. \’a0This (I think plausible) story is reinforced by another graph Bibow reproduces.\’a0 The global capital flow \’93paradox\’94 to which he refers in his title is the fact that developing countries are exporting capital to rich countries.\’a0 According to his data, developing countries have almost always been net recipients of private capital flows \’96 which is what one would have expected from most economic theory and history.\’a0\’a0\’a0

They have generally been net providers of official capital as far as foreign currency reserve accumulation goes, but for most of the last fifty years reserve accumulation on average was significantly less than net private inflows, so developing countries were net recipients of capital.\’a0 (For much of the 1980s the balance on both was zero or close to zero, and I suspect that this reflects negative private flows to Latin American and others among the 32 defaulted or restructuring LDCs, as they were then called, netted against positive private flows to Asia.)\’a0

It is only in 1998 that reserve accumulation among developing countries begins to take off and by 1999 it exceeds net private capital flows to developing countries.\’a0 This is when the \’93paradox\’94 of net capital flows from developing to developed countries begins.\’a0 Except for a small decline in 2001 net flows from developing countries surge almost in a straight line to around $700 billion annually (combining $1.2 trillion of reserve accumulation versus $0.5 trillion of net private flows).\’a0

I am sure there can be other competing explanations for the timing of these flows, but I am very impressed by the fact that Asian savings, as expressed in reserve accumulation, surge after 1997, as does the US trade deficit, although exacerbated by the second surge around 2004.\’a0 Given the virulence of the 1997 crisis and the tremendous shock it provided to Asian policy-makers (and policy-makers in developing countries elsewhere), it seems to me that a very plausible argument can be made that it was the effect of 1997 that caused the shift in developing-country policies that led to the surge in savings and the corresponding increase both in trade surpluses and reserve accumulation.\’a0\’a0\’a0

The surge in the US trade deficit after 1997 is also more easily explained by a shift in Asian trade policies and currency regimes than by a shift in US consumer preferences.\’a0 Of course that doesn\’92t mean that nothing relevant happened in the US.\’a0 US monetary policy was clearly too accommodative, and especially in reaction to the Iraq war, so that it exacerbated the conditions created by the Asian savings glut.\’a0 If anyone is still looking for which country to blame, my understanding of the creation of the imbalances suggests that you can blame almost anyone you like and there is a good chance that you\’92ll be at least partly right.\’a0

Why does this matter?\’a0

The issue of what drove what is not simply of academic interest.\’a0 The consequences for the world of a system in which imbalances were driven by a sudden and self-perpetuating explosion in US consumption, which then forced higher savings onto Asian countries, are very different from a system in which imbalances were driven by a sudden and self-perpetuating increase in Asian savings, which then forced higher consumption onto non-Asian countries.\’a0\’a0\’a0

Deciding whether or not the savings glut hypothesis is correct is important not just because it allows us finally to decide which country really is the evil predator, the US or China.\’a0 It matters for a very different reason.\’a0

If it was an explosion in US consumption which drove the global imbalances, then we are likely to see a fairly benign resolution to the crisis for everyone, except maybe the US.\’a0 After all in that case the imbalances were driven by US consumption excesses, and since those excesses are, like it or not, going to be resolved by the need for US households to repair their badly-damaged balance sheets, the imbalances too will be resolved, and in a way that is mostly benign for everyone except recovering US households.\’a0 This process may be postponed by current US fiscal policy, and especially by recent policies that subsidize consumption, but it will only be postponed, not derailed.\’a0

And just as Americans can no longer binge consume, their binge consumption will no longer force Asians to save such a high and rising portion of their income.\’a0 Asian growth, and especially Chinese growth, will be much more balanced.\’a0

But if the global imbalances were driven by a surge in Asian savings, Asian and Chinese growth will still rebalance, but the rebalancing will be much more difficult.\’a0 Why?\’a0 Because too-high Asian savings, caused in large part by post 1997 policies that encouraged differential growth between consumption and production (as I discuss here, for example), have until now been matched by too-low US savings rates.\’a0 As long as the two imbalances balanced, the world economy could continue functioning without too much distortion.\’a0

But now if we can expect net savings in the US (and perhaps in many other parts off the world) to rise, we need to see a rapid change in those policies that encouraged too-high Asian, and especially Chinese savings. \’a0In that light there was an interesting and worrying OpEd article in today\’92s Financial Times by the Peterson Institute\’92s Fred Bergsten and Arvind Subramanian:\’a0

The Obama administration is increasingly signalling that the US will not continue to be the world\’92s consumer and importer of last resort. The clearest statements came last month from Larry Summers , White House economics director, in a speech at the Peterson Institute for International Economics and in an interview with the Financial Times. The US, he said, must become an export-oriented rather than a consumption-based economy and must rely on real engineering rather than financial wizardry. Tim Geithner, the US Treasury secretary, and other top officials have spoken similarly of rebalancing US growth.\’a0

If the US really is serious about this shift towards higher savings, and if the primary source of the imbalance was the Asian savings glut, and not an original US consumption \’93glut\’94, this means that in the future US policies will be in direct conflict with still-current Asian policies, and unless the US is unable to accomplish these goals, Asian countries will need to force through an adjustment in their development policies as quickly as possibly.\’a0 Asian and especially Chinese officials have acknowledged the need to increase consumption more quickly.\’a0

But for now this adjustment in policies that encouraged too-high Asian, and especially Chinese, savings does not seem to be happening.\’a0 “The optimal choice is to expand household consumption,” PBoC governor Zhou Xiaochuan said in a speech last month. \’a0″That is, however, easier said than done. While the current income structure cannot be dramatically changed in the short term, the second-best choice is to maintain and expand investments.”\’a0 He is almost certainly right, at least except for his last statement.\’a0

In fact as I have argued many times (for example here, and here), I suspect that most of the Chinese fiscal stimulus is exacerbating the imbalances \’96 both by boosting current and future production and by creating conditions that will constrain future consumption growth.\’a0\’a0\’a0

In that case there has been no significant rebalancing yet towards more rapid consumption growth taking a greater share of Chinese production \’96 just a frenzied attempt to keep current growth rates high by boosting investment, which will almost certainly lead to capital misallocation and rising non-performing loans, and clearly unsustainable attempts by the Chinese government artificially (and unsustainably) to boost short-term consumption by subsidizing it heavily with government debt (something the US seems to have been doing too) which has the effective consequence of reclassifying fiscal expenditures as household consumption.\’a0

The end result?\’a0 Planned increases in investment in China eventually become forced increases in investment \’96 rising inventory \’96 that ultimately must lead either to writing inventory off or closing down production facilities in the future.\’a0 This is, by the way, just another way of stating the excess capacity problem.\’a0\’a0

Perhaps what we need is a real return to Confucian roots. \’a0I recently read this quote from Lao-Tzu: \’93The sage does not hoard. Having bestowed all he has on others, he has yet more. Having given all he has to others, he is richer still.\’94\’a0

Tags: Bibow

Predicting consumption growth in China

August 15, 2009 in Consumption and production by Michael Pettis | 31 comments

My ten-day trip \’96 partly holiday on Phuket and Phi Phi islands (even more beautiful than I had expected) and mostly meetings in Bangkok, Hong Kong and Singapore \’96 finally ended yesterday.\’a0 Besides three presentations to large investment groups, I met about twenty to thirty institutional investors in small meetings, and those meetings were very instructive.\’a0 There is a real mix out there it seems to me of tentative optimism about Chinese prospects on the part of the majority of investors and deep pessimism on the part of a minority, which included both Chinese nationals and foreigners.\’a0\’a0\’a0

Perhaps this is because of my own prejudices as a former bond-markets trader, but the pessimistic minority seemed much more experienced and literate to me.\’a0 They were also generally a lot more senior.\’a0 That might not mean much but it does suggest to me that there is a risk of bad news in the future causing a stampede of pessimism.\’a0 Of course the majority is not necessarily wrong (in spite of the claims of contrariarans, who paradoxically enough include nearly everyone, it seems), but the volatility impact of information that confounds their expectations is much greater than that of information that reinforces their expectations.\’a0

I may discuss the impressions I got from some of these meetings more later, but I wanted to talk a little about consumption growth in China.\’a0 My Financial Times OpEd piece published two weeks ago about the consumption constraint on Chinese growth seems to have generated several letters in disagreement.\’a0 I don\’92t think most of them made much sense as far as I could understand.\’a0 One seemed to have an overly optimistic assessment of the Chinese banking system.\’a0 It seemed to claim that since officially reported NPL ratios today are much lower than unofficial private estimates of ten years ago, the banks are in very strong shape, and so we don\’92t have to worry about the NPL effect of the recent surge in lending. \’a0That seems to me to be a defensible if illogical claim, but pretty implausible.\’a0\’a0

The letters also generally assumed \’96 and this confusion seems to happen an awful lot \’96 that the accounting identity GDP = C + I + G is something prescriptive, with GDP growth largely independent of consumption growth and sustainable in the medium and long term by forever-increasing amounts of government-financed investment.\’a0 Apparently the argument that foreigners won\’92t consume this rising current and future capacity, and that Chinese won\’92t consume this rising capacity either, which must lead either to forever-rising inventories, to dumping and trade tensions, or to write-offs, none of which is a happy solution, can simply be resolved by repeating GDP = C + I + G several times.\’a0

Nonetheless one of the letters in disagreement, by Professor Peter Williamson of Cambridge University made a lot more sense and identified the main weakness of my argument, as I see it.\’a0 According to Williamson:\’a0

Michael Pettis is right to remind us that correcting the fundamental imbalances in the global economy means that China’s export-orientated growth model is no longer tenable (\’93Get ready for lower Chinese growth\’94, July 31). But he is unduly pessimistic about the potential of Chinese consumers to generate sustained, double-digit growth in Chinese gross domestic product.\’a0

Between now and 2025 some 350m Chinese are set to become new city dwellers. This alone will give a boost to demand equivalent to the size of another Germany. The government\’92s stimulus package will also initiate more spending by consumers. The \’a373bn to provide universal cover for basic healthcare in China by 2011, for example, will encourage savers fearing medical bills to spend more freely. Workers made redundant from export-processing jobs, meanwhile, are adapting quickly \’96 finding new work and retraining (4m in Guangdong alone). Their consumption will rebound.\’a0

Local consumption is, indeed, the key to China sustaining high growth over the next decade. But that should be cause for optimism rather than despair.\’a0

Professor Williamson accepts the argument that China\’92s medium to long term growth rate will be constrained by the growth rate of domestic consumption, but he disagrees with me in assuming that the rate of growth in Chinese consumption (roughly 8-9% annually over the past several years) will remain the same or even decline in the next few years.\’a0 He thinks it is likely to grow much faster.\’a0

He may be right, and whether or not he is right is I think the key to the validity of my argument as a prediction of the future.\’a0 But I am not convinced.\’a0 For reasons I have discussed many times, I don\’92t think by his first argument, that China\’92s plan \’93to provide universal cover for basic healthcare in China by 2011,\’94 will cause a future explosion in consumption \’96 at least not in the time needed to address the adjustment in the global imbalances.\’a0\’a0\’a0

For one, simply because the government has announced the plan to put into place a good healthcare system by 2011 doesn\’92t mean that this is a done deal.\’a0 The government also announced several years ago that Chinese growth would become much more environmentally friendly, Chinese income would become more fairly distributed, corruption would be reduced, the banking system would be cleaned up, SMEs were to receive a growing share of loans, and the corporate bond markets would become a major source of funding, and yet anyone living in China might wonder if any progress had been made on any of those things.\’a0 Putting in a good health care system is very difficult in the best of circumstances, and with almost no accountability in local governments and local hospitals, and with rapidly rising deficits at both the central and provincial levels, I am pretty skeptical about their ability to achieve anything close to their stated ambitions.\’a0

But more importantly, even if they do pull it off, and give China Swedish levels of health care within a year or two, we shouldn\’92t assume that it will immediately affect consumption levels.\’a0 Anyone living in China knows how skeptical many Chinese are about government promises, and I would guess that it will take many years of testing the system successfully before even a significant fraction of Chinese household feel that it is no longer necessary to save enough to protect themselves from medical emergencies.\’a0 After all there are not many responsible adults who will throw caution to the winds when it comes to the possibility of an aged parent or, even worse, a son or daughter, needing serious medical treatment, and cash for that treatment.\’a0 In that case you don\’92t give up your safety net of hoarded cash until you are thoroughly and totally convinced that you will no longer need it, and that conviction only comes through experience.\’a0

I am also skeptical about his other claim for a surge in consumption based on a flexible work force quickly replacing export jobs with other jobs.\’a0 Unemployment seems to be rising (of course the official urban employment data is nearly worthless, but the anecdotal evidence isn\’92t good), and the main reason it hasn\’92t surged is probably because of the massive stimulus package, so once again we circle back to the argument about the sustainability of the stimulus package.\’a0 In that light let me just highlight a passage from the ft.com’s Dragonbeat by my friend Arthur Kroeber, a very smart China watcher with whom I used to disagree in the past a lot more than I do currently (perhaps a little to my chagrin, because given how much he knows about China, debates and disagreements with Arthur were always very educational for me):\’a0

China’s ability to maintain economic growth of around 8 per cent despite the global shock took many by surprise. But this ability has nothing to do with systemic advantages, a distinct \’93China model\’94 of growth, or skill in macroeconomic management.\’a0

Still less has it anything to do with the reasons cited by the People\’92s Daily editorial [Note: this is in reference to an especially silly editorial you can find here].\’a0 China\’92s present economic vitality results from a Great Wall all right \’96 a Great Wall of borrowed cash. There is nothing remarkable or spiritual about an economy growing at 8 per cent when credit is allowed to expand by 34 per cent.\’a0

The fact becomes even less remarkable when we recognise that nominal GDP (the appropriate comparator for nominal credit growth) grew just 3.8 per cent in the first half.\’a0 In other words, 10 dollars of new loans were required to generate just one dollar of economic growth.\’a0

In fact China\’92s first-half growth shows one thing and one thing only: the existence of a powerful state with the ability to commandeer its citizens\’92 wealth and plough it into more buildings, bridges and roads, with no regard for the return those investments will bring.\’a0

Not to be outdone, Shen Minggao, Caijing\’92s chief economist, also worries about the Chinese adjustment in an article in this week\’92s Caijing.\’a0 In particular he is concerned that Chinese consumption needs to accomplish the very difficult task of increasing at a rate commensurate with the increase in US savings:\’a0

During the S&ED the United States highlighted signs of healthy economic adjustments. For example, the savings rate for U.S. residents climbed to over 7 percent in June and will likely reach 10 percent in the future. The United States has done better than China in terms of these adjustments. China has to change its current exports-driven growth mode because a U.S. recovery does not necessarily boost consumption, given the soaring saving rates.\’a0

\’85The wealth decrease for American households did not shake the foundations of consumption.\’a0 Though the recession led to negative consumption growth, the largest year-on-year decline was only 4.4 percent, recorded in the last quarter of 2008. It was far less than the 8.9 percent drop during the late 1970s oil crisis.\’a0\’a0\’a0

However, it is uncertain whether deleverage of household balance sheets and rising savings rates are a one-time thing or a permanent change. As U.S. financial regulators tighten systemic risk supervision and easy lending becomes no longer available, American consumers have to rely on savings rather than borrowing to spend.\’a0

\’85In contrast, China’s high saving rates and constant inflows of foreign funding compel the Chinese government to make a hard choice: continue investing in infrastructure or buy Treasury bonds? China will have to suffer additional risks if the dollar depreciates but to continue investing in infrastructure will bring about low efficiency caused by overinvesting.\’a0

If China wants to avoid such a dilemma, it must boost domestic consumption. Chinese consumers must lower their savings rates. Meanwhile, the percentage of residents’ incomes in the gross national product should be increased greatly. Second, China needs to push for the appreciation of the yuan. Chinese firms have to rely more on domestic markets than overseas markets. Finally, China needs to turn potential consumption into real consumption during its industrialization and urbanization. Only in this way can China become a new engine for global economic growth and gain a bargaining advantage in future U.S.-China negotiations, which is commensurate with China’s economic size.\’a0\’a0\’a0\’a0

In another article another writer in the same magazine warns of growing overcapacity in steel:\’a0

The government will freeze approvals of new steel projects for the next three years in a bid to curb overcapacity, Ministry of Industry and Information Technology chief Li Yizhong told a news conference August 13.\’a0 Li said oversupply is a serious problem, with annual production capacity of 660 million tons far exceeding estimated demand of 470 million tons. China will produce a record 580 million tons of prime steel in 2009, far above the government’s target of 460 million tons, according to China Iron and Steel Association data.\’a0

So it\’a0mostly boils down to what policymakers can do to boost consumption growth \’96 net consumption growth \’96 in the short and medium\’a0term, and the sustainability of the fiscal stimulus to get the economy over the hump, and there is a whole lot of skepticism on my part, and on the part of many of my favorite China-based\’a0observers, that this is not going to be easy.\’a0\’a0\’a0

Of course Williamson may be right that there is a possibility of a surge in Chinese consumption, and if he is right my entire argument about a sharp slowdown in Chinese growth over the next decade will thankfully get thrown out the window. That is the main point \’96 my whole argument rests, I think, on whether or not consumption growth can be positively and, more importantly, sustainably affected by the current fiscal stimulus.\’a0 It is hard enough\’a0to present one’s full case\’a0in a letter to the editor, so I may be slighting the full extent of Williamson’s argument, but for now I think the evidence is still far more bearish than bullish on Chinese consumption growth.\’a0

What should have been discussed during the SED meetings (Part 2)

August 10, 2009 in Asian development model, Balance of payments, Consumption and production, Exports and imports, Global liquidity by Michael Pettis | 54 comments

In my last entry I tried to set out the necessary shifts over the next few years as the world, and especially China and the US, works out its imbalances. \’a0These shifts will take place, I am pretty sure, but they can do so under a \’93good\’94 scenario and a \’93bad\’94 scenario.

So what does all this have to do with the SED?\’a0 It means that the best hope for the two countries, I think, is a well coordinated set of policies acknowledging that the US savings rate must rise, and with it the Chinese must decline, but also recognizing that if this happens too quickly, or is accompanied by a collapse in trade, it will be bad for the US and terrible for China.\’a0 These coordinated policies must also acknowledge \’96 and this becomes much more difficult \’96 that the current Chinese stimulus may be making the adjustment more difficult, and much of it will have to reversed at the same time as the \’93appropriate\’94 measures aimed at spurring consumption may cause a short-term rise in unemployment.
\

\

Finally, the while the US commits to keep fiscal spending high, to turn a blind eye to trade disputes, and to run large trade deficits for several years more, China must commit to the financial sector and currency liberalization that will effectively reduce subsidies to producers and constraints on consumption.\’a0 The SED might also discuss the ability of workers to demand and enforce wage increases, since there is a wide consensus in China and abroad that among the main reasons for low household consumption in China is that wages are rising too slowly relative to GDP, and household savings are “taxed’ too heavily via interest rate policies.\’a0 Of course discussing workers right in a bilateral context is politically difficult, even without the irony of this particular discussion, so it will probably not happen.
\

\

When I discuss these issues, I am often confronted by the \’93aha!\’94 crowd who point out that my analysis must be wrong because if China does what I think they should do that would cause a rise in unemployment.\’a0 How can a policy be the right one if its implementation leads to a bad outcome?

That\’92s easy.\’a0 It can be the right policy if the alternative leads to a worse outcome.\’a0 That\’92s the problem.\’a0 There is no silver bullet here that can kill all the demons and leave us living happily ever after.\’a0 As I see it, the imbalances of the past decade were real and must be addressed, and we have broadly speaking three possible ranges of outcomes:

1. The US returns to its consumption orgy, the US trade deficit surges, and we\’92re back to the wonderful days of 2005. China can continue pumping out production and funding US consumption.\’a0 The problem of course is that this cannot be a permanent solution.\’a0 It just postpones the resolution of the global imbalances while fueling another asset bubble and saddling the US with even more debt and China with even more excess capacity.

2. China begins a long \’96 five or six years at least \’96 process of forcing the necessary structural changes that will permit a shift from a production-led economy to a consumption-led economy.\’a0 The changes necessary involve liberalizing interest rates and the banking system, allowing workers higher wages, and a number of other measures to boost SMEs, the service sector, and household consumption.\’a0 In the short term, however, nearly all of these measures will involve closing down unprofitable production facilities.\’a0 This must be done in conjunction with the US, so that the US adjustment is slowed down to a pace which China can absorb.\’a0 The US would do this by keeping fiscal expansion high enough to counteract the contraction in US household consumption.

3. Everyone does what they want to do anyway with no attempt at serious coordination.\’a0 US savings rise.\’a0 Chinese production rises too.\’a0 These two forces are globally incompatible and eventually lead to a sharp contraction in global GDP growth. \’a0The effects on China might include, but are not limited to, an explosion in Chinese inventory, a sharp and nasty contraction in international trade, or a brutal rise in Chinese NPLs and an unsustainable government debt burden.

High savings in China is not an accident.\’a0 Chinese trade and industrial policies that were aimed at generating employment growth by directly or indirectly subsidizing the cost of production, including currency and interest rate policies, nearly all effectively created forms of income and consumption taxes that constrain consumption even as they boost production (and a rising savings rates just means that production is growing faster than consumption), and to remove the latter you need to remove the former too.

It\’92s not so easy to increase consumption

So they have a dilemma: Remove the producer subsidies so as to allow consumption to grow, but cause subsidized producers to go out of business.\’a0 Or keep them in place, and perpetuate the production/consumption imbalance.

One way or the other Chinese policymakers are destined to be “successful” in raising the consumption share of GDP, because as the US reverses its earlier relationship between consumption growth and production growth, the rest of the world, which ran the opposite position, must also ultimately reverse.

Now for the next few years China’s savings rate will almost certainly decline and its consumption rate rise \’96 it has no other choice except to inflate a major, debt-fueled overinvestment boom \’96 but will that happen because of high growth in consumption or low production growth?\’a0 That is where policy matters very much, and the longer they wait to address the imbalance, the worse the outcome gets, I think.

Clearly Beijing wants to raise consumption quickly. \’a0Not too long ago a group government economists were reported to have reported on their website (sorry, but I lost the link):\’a0 “The new policy measures and initiatives will be the latest effort to shift growth from focusing on capital investment to a more sustainable model that gives domestic consumption a more important role in boosting economic growth.”

But they’ve been wanting to do this for a several years \’96 as they explicitly acknowledge by calling this the “latest” effort \’96 but the fact that it is harder to this now then it might have been three or four years ago doesn’t inspire me with much confidence.\’a0 It seems to me that most policies that will boost consumption in a stable and efficient way fall into one of two camps.\’a0 Measures like building the medical and social safety net, gradually getting banks to direct lending to service industries,\’a0 loosening the one child policy, and so on can be very successful, but will take years before they have much impact on real consumption.

In that camp I might add measures to force banks to increase consumer lending, because I think the last time they tried that (with car loans), nearly half the loans went NPL, suggesting that at first consumer lending will simply consist of free consumption financed indirectly by the government, when it bails out the NPLs.\’a0 This is a form of “consumption” I guess, but it is not really what the doctor had ordered.

Bad or worse
\

\

On the other hand reversing the policies that might have repressed consumption in the past will probably work more effectively within a shorter time horizon.\’a0 These would include liberalizing interest rates and allowing them to rise (which reverses the implicit transfer from households to producers), allowing workers to organize to demand higher wages, raising the value of the RMB, and so on.\’a0 Unfortunately nearly all of these measures would hurt manufacturers, especially in the export sector, and would cause an initial rise in unemployment.\’a0 I am not sure it is possible to manage the transition without a sharp, short-term rise in unemployment caused by the downsizing of the export sector as its implicit subsidies are removed, and it isn’t clear to me that any country that has managed a similar transition has been able to avoid this. My guess is China will have to do this, but will wait until they have no choice \’96 building up in the mean time even more excess capacity and bad debt. And bad debt, as I have argued before, must be resolved at some point in the future, and unfortunately usually in a way that constrains consumption growth.

One of the things that worries me is that the trajectory of rising US savings and increased investment in Chinese production is likely to squeeze the tradable goods sector in most countries around the world as China increase its market share. \’a0This will lead to accusations that China is behaving in a predatory way, and will almost certainly lead to increased trade tensions as policymakers around the world try to protect their tradable goods sectors form \’93unfair\’94 Chinese competition.

But I don\’92t believe that China should be considered predatory. China desperately wants to raise its consumption rate, because it is highly likely that for the next few years Chinese GDP growth will be limited to something below Chinese consumption growth.\’a0 Beijing would love to find the magic policy that transforms Chinese consumption overnight and turns China into a continental economy driven by internal demand.\’a0 It would love to see the trade surplus reduced not by a collapse in exports but rather by a shifting of exports to domestic consumption and a rise in imports (this last maybe).
\

\

The problem is that there is no such magic policy.\’a0 I cannot find any historical precedent of a country that was able to make the transition quickly and painlessly, and because of its own domestic problems \’96 especially the employment effect of the contraction in the export sector \’96 China is facing a difficult set of policy choices. \’a0The fact that the fiscal stimulus may be exacerbating China\’92s reliance on the export sector was not the plan. \’a0The fiscal stimulus is aimed at arresting a sharp and probably politically unacceptable rise in unemployment, and the fact that so much spending has gone into investment, rather than consumption, reflects rigidities in the economic and financial structure. \’a0China would love to see explosive growth in domestic consumption, but there is no way they can easily engineer such growth.

So we are stuck with policymakers, in China and elsewhere, making the best of a bad situation. \’a0They can be criticized for not beginning the adjustment process when conditions were much easier, but that is a criticism that can be spread around pretty thickly to policymakers in quite a few countries.\’a0 Anyway it is too late.

In these circumstances policy coordination matters a lot, and I see too little of it to have much optimism.\’a0 Beijing, Washington and Brussels must recognize that China and the world is still in a more vulnerable position than anyone seems to realize, and that rising US savings and rising Chinese investment create conditions for two seemingly irresistible forces to go head to head, and without coordination the consequences could be much worse than we expect.

What should have been discussed during the SED meetings (Part 1)

August 7, 2009 in Asian development model, Consumption and production, Economic growth by Michael Pettis | 34 comments

By coincidence I had two OpEd pieces that came out last week, one in the WSJ and the other in the Financial Times.\’a0 The latter came about because about a month ago Martin Wolf asked me to write a piece based on my June 20 entry.\’a0 The former came about on the previous Friday when I was thinking about last week\’92s SED meeting and why I wasn\’92t expecting much to come from it.\’a0 Although they are very different pieces, both of them build on this idea that the inversion of the consumption/GDP growth relationship in the US has important implications for China\’92s future GDP growth.\’a0

For the WSJ piece I start by pointing out that when the Japanese and German currencies soared in value against the dollar after the Plaza Accords were signed in September 1985, many analysts thought that at long last their trade surpluses with the US would decline.\’a0 They were partly right in the sense that the German trade surplus with the US did indeed decline.\’a0 But in spite of the fact that the value of the yen doubled, Japan\’92s trade surplus nonetheless surged.\’a0

I don\’92t think this should have come as a surprise.\’a0 There is a tendency to think that the value of the currency and the level of import and export tariffs are the main policy tools affecting the trade balance, and so absent a change in tariffs, any increase in the value of a country\’92s currency will automatically lead to a decline in its trade surplus.\’a0\’a0\’a0

Trade surplus

In fact the trade surplus reflects the gap between what a country produces and what it consumes, and so anything that affects that gap is implicitly a trade policy.\’a0 I discussed this in some depth in my June 3rd entry.\’a0

In the case of Japan in the post-Plaza Accords environment, the Ministry of Finance and the Bank of Japan responded to the currency agreement by directing a flood of low-interest credit into the manufacturing sector while informally guaranteeing borrowers, so assuring lenders that profitability was irrelevant in determining the flow of credit.\’a0 Sound familiar?\’a0 As a consequence Japanese manufacturers increased their production even as the flow of funding into the manufacturing sector and traditional constraints on household consumption forced an increase in the gap between Japanese production and Japanese consumption.\’a0 The result: a rising trade surplus.\’a0\’a0\’a0

By the way I have been reading Akio Mikuni and R. Taggart Murphy\’92s Japan’s Policy Trap: Dollars, Deflation, and the Crisis of Japanese Finance, an interesting book that covers a lot of this ground.\’a0 I recommend it to China watchers, although I am no expert on Japan and I did have a big problem with the often-repeated assertion (and one that often pops up in discussions\’a0about China) that because Japanese trade was not denominated in yen the Bank of Japan was forced to accumulate dollars.\’a0 In fact it doesn\’92t matter what currency your trade is denominated in \’96 if you run a net current and capital account surplus, your central bank must accumulate foreign currency.\’a0 Had trade been denominated in yen foreign buyers would still have had to convert dollars to yen with the Bank of Japan in order to make their purchases.\’a0

But that is a digression, and aside from a few irrelevant disagreements I think the book is quite illuminating.\’a0 In China, like in Japan during the 1980s, there are a number of factors besides the value of the currency that affect the country\’92s trade account, and even if the value of the Chinese yuan rises, it will not automatically lead to a decline in China\’92s trade surplus commensurate with the contraction in global trade, especially if it is matched by a significant credit expansion to the manufacturing sector.\’a0

Several policies are aimed at boosting production besides the undervalued currency.\’a0 As I have discussed before, these include very low lending rates enforced by the People\’92s Bank of China, energy and commodity subsidies, and probably most importantly, a flood of credit aimed at investment both in infrastructure and in the manufacturing sector.\’a0 At the same time very low deposit rates, constraints on consumer financing, and low wages, among other factors, prevent consumption from growing at nearly the pace necessary to absorb everything that China produces.\’a0\’a0\’a0

As an aside MacQuarie\’92s Paul Cavey has a very interesting OpEd piece in last week\’92s Wall Street Journal, based on a longer research piece which I am not able to link.\’a0 Among other things he argues that although China has run negative interest rates for much of recent history, until last year there was no credit bubble because credit was rationed \’96 and credit rationing implicitly raises the cost of capital for the system, even if interest rates are nominally low. \’a0Recent conditions, however, are different, and all rationing has disappeared with the explosion in credit of the past eight months.\’a0\’a0Cavey concludes:\’a0

It\’92s not impossible for Beijing to take away the punch bowl of credit. There is plenty of room to defy the skeptics and in the next few months and push through structural reforms. For instance, some of the privileges state-owned enterprises continue to enjoy in terms of the ability to provide domestic services like banking and telecoms could be dismantled, allowing the country\’92s more productive private sector to thrive in local markets rather than just overseas. But without such changes China will be relying on growth financed by cheap domestic debt. This means China will be decoupling itself from the U.S. consumer, but at the cost of a credit bubble.\’a0

China\’92s consumption will rise\’a0

So to return to the main story, with the credit expansion and other measures aimed at boosting production, will China\’92s trade surplus soar?\’a0 Probably not.\’a0 Every trade surplus requires a trade deficit elsewhere, and as the leading trade deficit country, policies in the US that affect the gap between consumption and production will also determine the size of the US trade deficit.\’a0 If the Obama administration is successful in forcing a rise in US savings levels, and even if it is not (since in the short term US households have no choice but to increase their savings rates), US consumption must grow more slowly than US production and the US trade deficit will narrow, except in the very unlikely case that US investment soars \’96 investment would have to grow faster than savings to keep the trade deficit from contracting.\’a0

For China this almost certainly forces the country into either of these two outcomes\’a0\’a0

1.\’a0 The government continues the current fiscal expansion forever, in which a huge expansion in government-led investment pushes growth forward.\’a0\’a0\’a0

2.\’a0 The consumption rate in China must rise as a share of GDP.\’a0

There are at least three problems with the first option.\’a0 First, a significant portion of the fiscal stimulus (and almost certainly a higher share than reported) is directed into manufacturing in the tradable goods sector, which needs anyway to be absorbed by rising consumption, either in China or globally.\’a0 Second, given the inefficiency of the current fiscal and credit expansion, and the concomitant rapidly rising direct and contingent government debt, there is a real question as to whether this program can be sustained for more than one or two years.\’a0\’a0\’a0

And third, and this seems to be the most confusing point for some, the economic purpose of investment is to increase future production, and even if the fiscal stimulus turns out to be hugely efficient (it isn’t), without a surge in future domestic consumption to absorb the additional Chinese capacity we will still be stuck with the need for a massive return to US profligacy, and Chinese funding of that profligacy, to absorb the increased production.\’a0\’a0

The first option, in other words, is at best possible for a very short time, and ultimately we are forced into the second option: Chinese consumption must rise as a share of GDP, or to put it another way, Chinese GDP must grow more slowly than consumption.\’a0\’a0\’a0

So why should the US care what China does to rebalance its trade if changes in US consumption will force a rebalancing anyway?\’a0 Isn\’92t discussion and coordination pretty much unnecessary if a rising savings rate in the US must ultimately force an adjustment on China?\’a0\’a0\’a0

No.\’a0 US and Chinese policies matter because there are many ways that international trade can rebalance.\’a0 In the US we will see consumption grow more slowly that production, just as in China we will see consumption growth outpace production growth.\’a0\’a0\’a0

Both will happen, but in both countries there is a good scenario and a bad scenario.\’a0 The good scenario for the US would see some growth in consumption buttressing healthier GDP growth.\’a0 But the bad scenario would involve a contraction in GDP driven by even faster contraction in consumption.\’a0 For China a good scenario would involve surging consumption driving slightly slower GDP growth, and a bad scenario would consist of slow consumption growth dragging down GDP growth.\’a0\’a0\’a0

If China continues to pump out capacity and tries to export this excess abroad, and if US household savings rise much more quickly than US fiscal dis-saving (borrowing), we will almost certainly see the bad case scenario occur, at least in China, and especially if it leads to trade friction around the world.\’a0 The nightmare scenario is that in the US a still-high trade deficit prevents a slowdown in consumption from nonetheless causing a sharp slowdown in economic growth, which leads to rising unemployment, which causes consumption to slow down even further.\’a0 Meanwhile in China rising inventories eventually lead to cutbacks in production, which also lead to rising unemployment.\’a0

As fewer Chinese get jobs, the unemployed consume less, and the employed also try to increase their savings because of rising uncertainty.\’a0 Since net Chinese savings must decline if net US savings rise (note I am assuming the rest of the world, including sustained investment levels, is constant, but I suspect the impact of the rest of the world will actually be adverse), the only way for this to happen if the Chinese savings rates rises is either for a burst of inefficient and unsustainable debt-fueled investment by the government, or for GDP growth actually to slow sharply.\’a0

I know all this sounds drastic, but the imbalances have to be worked out one way or the other.\’a0 Rising savings in one part of the world, even assuming no changein global investment, requires declining savings somewhere else, and although it may be unrealistic to expect no change in global investment, the plausible prediction is that global investment will actually decline, which increases the pressure.\’a0 This is just another way of saying that changes in trade deficits in one part of the world require equal changes in trade surpluses elsewhere.\’a0 This is also just the obverse of saying that declining consumption in one part of the world requires rising consumption elsewhere (or sharply rising investment, which since it represents future production only postpones the need for consumption growth) or else global GDP must contract.\’a0

Uncoordinated policies\’a0

What will determine whether or not the two countries follow the good scenario or the bad scenario?\’a0 Clearly fiscal and monetary policies in both countries will matter because they will set the speed of the adjustment and they may or may not speed up the adjustment process.\’a0

In the US, fiscal expansion is aimed primarily at slowing the pace of demand contraction.\’a0 This may be necessary since I expect US consumption will grow slower than US GDP for many years, but it comes at the expense of a rising fiscal debt.\’a0 I am not as worried as many others seem to be about US fiscal indebtedness and I am certainly not worried about the ability of the US to fund its debt, especially since the stock of debt in the US is declining (private debt is dropping faster than public is rising).\’a0 As I have argued many times, I also think all the fear-mongering about whether or not China and other foreigners will continue to fund the US fiscal deficit is totally muddled thinking and among the least important things to worry about.\’a0 Foreigners will and must fund the US current account deficit, and the bigger the deficit the more they will fund \’96 so really we actually want foreigner to reduce their funding.\’a0\’a0

But there are reasonable limits to how much debt we want to see in the US, and we certainly don\’92t want to see a continuation of the global imbalances in which the debt-fueled consumption binge of US households is simply replaced by the a debt-fueled consumption binge by the US government, especially since as long as the trade deficit is high a large part of the job-creating aspect of US fiscal deficits will leak abroad, requiring even larger US fiscal deficits.\’a0 In addition, the US fiscal program should be accompanied by specific measures aimed at increasing US household savings \’96 I am not able here to go into much detail on how to do this (and I am no expert on the subject), but for example perhaps we can eliminate taxes on interest income, raise consumption or gasoline taxes, and so on.\’a0

Of course forcing an increase in US savings means improving the long-term US outlook while hurting short-term prospects for employment.\’a0 Rising US savings means declining consumption growth, and remember that US GDP growth will be less than growth in US demand for the next few years as US debt levels decline.\’a0\’a0\’a0

I think China will face an even more drastic version of this trade-off, and this is because, as I have been arguing for two years, contractions in global demand force the most difficult adjustments not on the \’93sinful\’94 low-savings trade-deficit countries but rather on the \’93virtuous\’94 high-savings trade-surplus countries.\’a0 China needs to cut capacity drastically and put into place the factors that will lead to a rise in net consumption, but most of these policies will actually hurt employment in the short term.\’a0 I have already discussed what these policies are likely to be in my June 3rd entry, and almost all of them will almost by definition force a contraction in the tradable goods sector.\’a0

China\’92s problems will be made much worse if it is forced to cut capacity very quickly, which will happen if trade disputes get worse.\’a0 Already disputes with Asian neighbors are pretty nasty, and they are likely to get worse with the US and Europe.\’a0 There has been a lot of discussion recently about China turning to other developing countries as sources of net demand to replace the US, but this is unlikely.\’a0 Aside from the fact that no one is large enough, none has the ability to run persistent trade deficits.\’a0 China can fund these deficits for a while, but it will learn, as many have before it, that funding persistent current account deficits for developing countries eventually leads to defaults on the debt.\’a0

So after all the premable on what do I think the SED discussions should focus?\’a0 Since this entry is long enough already I will postpone that part of my discussion for a couple of days.\’a0

Squeezing out the exporters

July 29, 2009 in Balance of payments, Consumption and production, Exports and imports, Fiscal stimulus by Michael Pettis | 42 comments

I am working on a fairly long entry that I will post this weekend about why a trade rebalancing and a consumption/savings rebalancing will take place in both China and the US whether or not we want it.\’a0 This week has been crazy, among other reasons because a festival in Taiwan has invited one of our indie bands and one of our experimental bands (Carsick Cars and White) to perform this weekend at the Music Terminals Festival in Tao Yuan City.\’a0 Getting visas for these kids has been brutally difficult and they actually had to cancel one of their club gigs, on Thursday, because of problems with getting things done on time.\’a0 Still, if any of my readers are going to be in Taiwan this weekend, I strongly recommend that you check out the festival, which besides the two Beijing representatives features a lot of great bands from around the world (or if you prefer club gigs, check them out Friday night at a pre-festival show at The Underworld, in Taipei).

So much for the good news.\’a0 The bad news is described in an alarming article in today’s Wall Street Journal which shows that trade tensions are continuing to rise.\’a0\’a0

European Union trade officials approved pre-emptive penalties on imports of steel pipe from China, a precedent-setting move that suggests the trading bloc is growing more protectionist in the face of the economic downturn.\’a0 Tuesday’s vote by trade officials from the EU’s 27 member states is significant, say trade experts, because they accepted an argument from steel producers \’96 including the world’s largest by volume, ArcelorMittal \’96 that punitive tariffs are needed to protect them from the threat of underpriced imports from China.\’a0 Previously, complainants have had to prove the imports had already hurt their businesses. Trade lawyers say they expect a host of industries to ask the EU for protective tariffs in August.\’a0\’a0

I have been hearing rumblings for a while about tougher stances being taken in Europe and the US in response to the perception that China is exacerbating the global contraction in demand by increasing subsidized resources available to manufacturers, most importantly by channeling a huge increase in lending at interest rates subsidized by Chinese household consumers and socializing the risk.\’a0 These new protectionist moves seems to be an expression of just this.\’a0 The article goes on to say:\’a0\’a0

Basing a claim on the threat of injury “is a perfectly legal strategy, but it has simply not, until now, been used as a matter of EU policy,” says Nikolay Mizulin, a Brussels-based trade lawyer with Hogan & Hartson LLP. \’a0This case “is a sign of growing protectionism and could open the floodgates to many more industries who believe they deserve protection.”\’a0 Mr. Mizulin and other trade lawyers say they expect many industries to seek protective tariffs next month.\’a0\’a0

As I have been arguing for over a year, as unemployment around the world rises and as the necessary contraction in US net demand picks up pace, there was inevitably going to be a conflict with China as Chinese policymakers responded to the collapse in trade in the only way they could, by substantially stepping up investment.\’a0 The result is that China\’92s trade surplus has contracted very slowly \’96 much more slowly than the contraction in the US trade deficit \’96 and the result was a huge squeeze on the tradable goods sectors around the world.\’a0

The fact that policymakers in Europe, China, Japan and the US seem to have no clue as to how difficult the transition for each of the other countries is likely to be, and so are doing not nearly enough to coordinate their response (in fact lecturing and finger waggling seem to the favorite forms of policy coordination), makes trade conflict almost a dead certainty. \’a0I don\’92t think there are necessarily any bad guys here \’96 each country is desperately doing what it can to get itself out of this mess \’96 but there is a lot of failed opportunity and I am pretty sure that the trade environment will continue to decline.\’a0

The problem is illustrated in two interesting recent pieces. \’a0My friend Dan Rosen, of the Rhodium Group, has a very illuminating July 17 report that shows the composition of Chinese growth in the past decade.\’a0 He shows that for the past five years net exports accounted for about 10% to 15% of Chinese GDP growth, before collapsing\’a0to minus 41% in 2009 YTD.\’a0

Until recently investment’s share of GDP growth peaked at around 65% in 2003 \’96 a very high share by any standard \’96 and going back the full thirty years of China\’92s reform period achieved an historical high astonishing of 81% in 1985.\’a0 From 2005 to 2008 the investment share of GDP growth averaged around 40% \’96 still high \’96 and then in the first half of this year accounted for a mind-boggling 88% of this years GDP growth.\’a0

This year\’92s growth, in other words, is almost wholly a function of the massive increase in investment, and this increase in investment started out largely in the form of reopening production facilities and producing more \’93stuff\’94, without any significant rise in consumption.\’a0 As we know, when production increases faster than consumption, either the trade surplus or inventories must rise.\’a0

On that note Xinhua published the following article on Monday:\’a0

The per capita consumption spending volume of Chinese urban residents stood at 5,979 yuan (875 U.S. dollars) in the first half of this year, up 8.9 percent year on year, the National Bureau of Statistics (NBS) announced Monday. \’a0Deducting price factors, the growth reached 10.3 percent.\’a0\’a0 The per capita disposable income of Chinese city dwellers rose 9.8 percent year on year to 8,856 yuan in the first six months. Deducting price factors, the increase reached 11.2 percent, said the NBS.\’a0

Consumption has been rising at around 9% a year for the past several years.\’a0 Notice that if GDP growth slows to under 9%, the savings rate in China will automatically decline.\’a0

The second interesting piece is put out by the Economic Policy Institute, a group I believe not noted for its commitment to free trade. \’a0It shows China\’92s share of the US trade deficit excluding oil. \’a0According to their numbers:\’a0

Year

2000

2001

\
\’a0
\

2002

2003

2004

2005

2006

2007

2008

2009

Share

26%

27%

\
\’a0
\

28%

31%

35%

40%

45%

54%

69%

83%

Perhaps as a consequence of a fiscal stimulus aimed at boosting investment and production, China’s share of the US trade deficit has grown significantly. \’a0Since the US trade deficit is shrinking quickly, this means that other exporters are getting killed.\’a0 As I have argued for a while, this is not sustainable and will almost certainly cause trade tensions to erupt.\’a0

Does this mean China is behaving in a predatory way?\’a0 I don’t thinks so.\’a0 I have warned for a long time that it would be very difficult for China to make the necessary transition to a consumption-led economy quickly enough to accommodate the global adjustment taking place. \’a0Unless it is willing to see its economy collapse, there is simply no way China can reduce its negative net demand quickly enough to match the contraction in US demand and so avoid squeezing the hell out of the global tradable goods sectors.\’a0 That is why policy coordination is so important, especially between China and the USD, and of course that is why I continue to be a pessimist. \’a0I do not think this policy coordination is taking place.\’a0 I will write about this more later this week.\’a0

To continue the discussion of last week, we are getting more conflicting signals about policy confidence. \’a0On the one hand Bank of China seems to love this party. \’a0According to an article in today’s Bloomberg:\’a0

Bank of China Ltd., which doled out the most loans among Chinese banks in the first half, plans to keep expanding credit unless the government clamps down on the nation’s record lending boom.\’a0 The nation’s third-largest bank will maintain its original target of generating about 10 percent of China\’92s new loans in 2009, Beijing-based spokesman Wang Zhaowen said by telephone yesterday. Bank of China may \’93fine tune\’94 its strategy in line with any government policy changes, he said.\’a0\’a0

\’85Bank of China will continue to lend to 10 key industries with government policy support, including steel, shipbuilding and automobile, Wang said. About 30 percent of its loans went to those industries in the first half.\’a0\’a0

On the other hand two of the other members of the Big Four seem a lot more cautious. \’a0Today\’92s South China Morning Post has this article:\’a0

Mainland\’92s two biggest state-owned commercial banks have put a lid on their lending targets for the year, according to domestic media reports, in a move that will significantly slow overall credit growth in the second half. Industrial and Commercial Bank of China (ICBC) is aiming to issue full-year new loans of 1 trillion yuan (HK$1.3 trillion), while China Construction Bank (CCB) has set a goal of 900 billion yuan, Caijing magazine reported.\’a0

The two banks, mainland’s largest by market value, granted new loans of 825.5 billion yuan and 709 billion yuan, respectively, in the first half. \’a0If they stick to their reported targets, this would imply that ICBC would have already issued 83 per cent of its full-year lending total, while CCB would have already issued 79 per cent.\’a0

It is surprising to me that these members of the Big Four are responding so differently, at least in public. \’a0I wonder if the management of the different banks belong to different factions and so interpret the fiscal stimulus package differently. \’a0Perhaps my friend Victor Shih, who understand these things better than I do and who sometimes reads my blog, might comment?\’a0

Finally the Financial Times on Monday continued the thread discussed in my Saturday post with an article called \’93China warns banks over asset bubbles.\’94\’a0

Chinese regulators on Monday ordered banks to ensure unprecedented volumes of new loans are channelled into the real economy and not diverted into equity or real estate markets where officials say fresh asset bubbles are forming.\’a0 The new policy requires banks to monitor how their loans are spent and comes amid warnings that banks ignored basic lending standards in the first half of this year as they rushed to extend Rmb7,370bn in new loans, more than twice the amount lent in the same period a year earlier.\’a0

\’85Beijing\’92s concerns are echoed in other countries across the region, most notably South Korea, where the government says it is taking steps to cool a real estate bubble, and Vietnam, where the government has ordered state banks to cap new lending to head off inflation.\’a0 regulators are now concerned that too much money is being lent by the state-controlled banks and the country\’92s tentative economic rebound could come at the cost of a stable financial system.\’a0

In statements published last week, Wu Xiaoling, who recently retired as deputy governor of the central bank, warned new lending this year would probably reach as high as Rmb12,000bn, a staggering increase of 40 per cent of the entire stock of outstanding loans in just one year.\’a0\’a0

\’85Ms Wu hinted Beijing may soon raise the amount of money banks must hold on deposit with the central bank, marking a change of policy from last year when it aggressively slashed the reserve requirement ratio and interest rates.\’a0 The central bank has also ordered 10 banks, including Bank of China, to buy Rmb100bn worth of central bank notes with a maturity of one year and a return of just 1.5 per cent, according to Chinese media reports.\’a0 This move is interpreted as a warning to banks that have been the most active lenders that they should now start to rein in their excessive behaviour.\’a0

« Older entries

Consumption and production

The IMF warns about surplus countries and global imbalances

More trade tensions, and the very limited advantage of relative poverty

The coming clash in savings

China\’92s August data confirms both optimists and pessimists

It\’92s not yet the end of China\’92s massive stimulus

Yet another discussion on the Asian savings glut hypothesis, and why it matters

Predicting consumption growth in China

What should have been discussed during the SED meetings (Part 2)

What should have been discussed during the SED meetings (Part 1)

Squeezing out the exporters

Shows

Categories

Recent Posts

Recent Comments

September 2010
M	T	W	T	F	S	S
« Oct
		1	2	3	4	5
6	7	8	9	10	11	12
13	14	15	16	17	18	19
20	21	22	23	24	25	26
27	28	29	30