His title hangs loose about him, like a giant’s robe upon a dwarfish thing – William Shakespeare
The United Nations Population and Demographics Office (UNPDO) reports that the average height of adult Chinese women has reached 170 centimeters (5 ft 6.9 in), up from 159 centimeters (5 ft 2.6 in) in just about 25 years. On average, adult Chinese women are about four centimeters (1.57 in) taller than American women.
According to an opinion poll conducted among a group of American, European and Asian professionals who either have lived in or have visited both China and America, 97% attributed the increase in height to “better nutrition in the Chinese diet” or other factors such as more in inter-racial marriages as a result of China’s open door policy. Only 3% of those polled suspected that anything was wrong with the data.
I apologize to the respondents that I, the pollster, made up the Chinese height statistics and that the UNPDO does not exist. But I thank them for helping me confirm a hypothesis: People generally do not question statistics from what seem like authoritative sources, such as the United Nations, no matter how implausible the information is. Everyone who has traveled in both countries knows that Chinese women are not taller on average than American ones. And it is also impossible for half of the adult population of a nation of more than one billion people to grow its average height by as much as 11 centimeters (7.1 inches) in 25 years. This exercise proves, however, what many very smart people believe is right can still be totally wrong.
Such is the case, as I have argued in these pages and elsewhere, with some of the major findings of the World Bank on what return on capital China produces and on how Chinese firms finance their investments.[1] At issue is not whether the profits of Chinese firms have grown, as the World Bank researchers suggest. No one disputes that. But a growing dwarf is still no giant. At issue is whether China allocates and uses capital efficiently enough to produce a return on capital as good as or better than international markets. The question is: How efficient is China in using its capital by international standards?
My own analysis of the same data used by the World Bank concludes that the return-on-equity numbers reported by it are significantly overstated because they do not net out such items as corporate income taxes. The World Bank does not dispute this. But it insists that Chinese firms are now making so much profit that undistributed profits or retained earnings finance more than half of their investments whereas bank loans only one-third or as little as one-sixth. If this is true, it suggests that Chinese firms, on average, finance their investments with much less debt in proportion to equity on average than almost all their international peers and therefore, by implication, they must be more profitable.
For evidence, the World Bank points to the increase in the proportion of corporate savings in China’s national savings, now accounting for more than 20% of GDP.
Some believe that Chinese corporate savings are over-estimated and household savings under-estimated.[2] Nevertheless, assuming that these numbers are correct and that corporate savings do represent more than half of corporate investments (or whatever percentage the World Bank says, as it has used, rather loosely, different numbers in different contexts), does this mean that Chinese firms on average finance more than half of their investments with undistributed profits?
All Air Is Not Oxygen
The World Bank researchers argue that corporate savings, as a macro-economic concept, are equal to undistributed profits or retained earnings by definition. Unfortunately that definition is as correct as defining the air we breathe as pure oxygen. As a macroeconomic concept, corporate savings consist of more than just undistributed profits.
In addition, corporate savings also include depreciation, amortization (both of which are treated as costs in any corporate income statement) and other things (including, guess what, government subsidies). Depreciation alone is a big number. China’s growth is driven by fixed asset investments which now represent more than 50% of GDP and which are still growing at 25% to 30% per year. As such, China’s fixed asset base expands each year and so the amount of depreciation from such an asset base also increases. Nobody knows for sure how big the depreciation number is because China’s statistical authorities do not have complete data in this regard; but one can make some safe estimates. The enterprise data series published by National Bureau of Statistics include the net fixed asset value of industrial firms of scale.[3] By NBS estimates, depreciation represents about 12.5% of net fixed assets in 2003, up from 9.1% in 2000 and 2001 and 11.5% in 2002. Based on this number, depreciation is roughly 11% of net asset value for 2005. The net profit of these firms should be no more than 8% to -9% of net asset value in 2005[4]. But for the sake of argument, let’s use 9% and also assume there is no dividend payment (when in fact, many Chinese companies do pay dividends). By this calculation, depreciation alone (11% of net asset value) is more than 1.2 times as large as undistributed profits (9% of net asset value) or at least 55% of corporate savings. Undistributed profits at most represent about 45% of corporate savings.
Therefore, if corporate savings represent around 60% of corporate investments, as the World Bank suggests, undistributed profits account for no more than 27% of corporate investments, from a macroeconomic point of view. How remarkable a percentage is that? Not very. Importantly, as long as China’s fixed asset investment continues to grow at the current rate, then depreciation, and therefore corporate savings, will grow accordingly, regardless of whether there are any undistributed profits.
Other sources of data which the World Bank researchers claim to have checked against to support their story of investments mostly financed by undistributed profits do not provide a different picture. For example, in enterprise survey data, “self-possessed” or “self-raised” funds both include funds substantially larger than undistributed profits, such as depreciation, amortization and other items which no business calls undistributed profits.
Who Bears the Risk?
But this is only the beginning of what the World Bank experts got wrong. Now we know that corporate savings are not equal to undistributed profits. But assuming corporate savings are mathematically equal to 60% of corporate investments, can we therefore conclude, as the World Bank does, that corporate savings finance 60% of China’s corporate investments? The answer is still an emphatic no. In fact, there is absolutely no way anyone can draw any inference on how firms finance their investments in aggregate from corporate or national savings data. Even if the amount of corporate savings exceeds corporate investments, as in the United Sates, bank loans can still finance as much as 100% of corporate investments. The key to knowing who provides financing is to understand who bears the risk of providing money to enable investment.
A simple example will make this clear. Suppose there are only two firms, A and B, and one bank in a simple economy. Firm A, which has no debt, puts $1000 in savings into a bank and Firm B, which has no equity, borrows $1000 from the bank to make a $1000 investment. In this case, the World Bank would say that corporate savings finance 100% of corporate investment from a macro-economic standpoint. But what happens if the investment is completely lost? Do corporate savings disappear? They do not. Ultimately, it is the bank, not Firm A, which has provided financing for Firm B, because every time the bank makes a loan, it puts its own capital at risk. If the loan is lost, the bank suffers the loss but is still obliged to pay back Firm A. The role of the bank is far from simple intermediation, such as the role played by a brokerage firm selling stocks for its client. The bank is a provider of risk capital. This ought to be obvious to anyone who has ever had a bank account.
What if the bank does not have adequate capital to cover the losses? In most cases depositors will still get their money back either through a government sponsored deposit insurance scheme or, as in the case of China, a government bailout. Ultimately, of course, it is the taxpayers who will have to pay.
How do we know how much of corporate investment in China is financed by banks? There used to be no confusion before the World Bank came along. The data series on Chinese industrial firms provides information on the aggregate debt-to-equity ratio, which is the quotient of the total amount of debt divided by the total amount of net asset value. This ratio, at about 150%, has remained basically unchanged over the past five years, indicating that in aggregate, Chinese industrial firms, at least those in the NBS database, finance 60% of their investments with bank loans. Therefore, regardless of the fact that corporate savings are mathematically equal to 60% or 100% of corporate investments, bank loans still finance the majority of China’s corporate investments.
The actual aggregate debt-to-equity ratio must be even higher. This is because a large amount of corporate debt has been converted into equity in the past few years either through debt restructuring or after major banks removed their bad loans to government owned asset management companies. The total value of such transfers is about $432 billion, or about 18% of China’s 2006 GDP, according to Moody’s. The debt-to-equity ratio would certainly be significantly higher than reported if none of these bad loans had been converted into equity.
Those who argue that the aggregate debt-to-equity ratio has declined since 10 years ago are simply not familiar with China’s history of the reforms of the state-owned sector. The debt-to-equity ratio at that time is not comparable with the current numbers, as it was around then that a massive amount of debt had been converted into equity through a nationwide “debt to equity conversion program.” This happened long before the recent recapitalization of major Chinese banks. One of the companies I studied 10 years ago, Tianjin Seamless Steel Pipe, had about $1 billion debt on its balance sheet but no equity (bizarre as it might sound). Through the nationwide “debt to equity conversion program,” a large amount of debt was converted into equity, contributing to the sharp decline in debt-to-equity ratio at that time. This program was already history five years ago.
The reason that banks accumulate bad loans or suffer losses is because they over-lend or under-price their risks. This allows the borrowing firms to over-invest or invest in projects which do not produce returns sufficient to cover their risks. Therefore, the amount of bad loans in the banking system is the best indicator of inefficiencies in an economy. If on average, return on capital is sufficient to cover investment risks, there should not be significant amount of bad loans. How much have Chinese banks over-lent or mis-priced their risks? Moody’s estimates that it would require $620 billion, which is equivalent to 25% of China’s 2006 GDP, to recapitalize China’s banking system, including the $432 billion already carved out of the banks but still largely unfunded by the government.[5] These numbers are largely based on official accounts and reports. As such they are very conservative. Other independent estimates have put the capital required much greater. Regardless, to put this number into perspective, it is roughly equal to the combined pre-tax profits made by all the industrial firms in the NBS database in the past seven years!
The World Bank argues that Chinese banks’ nonperforming loan ratio has fallen in recent years. That is true. But to date, it is the government sponsored removal of bad loans from major state-owned banks, the capital injection by the government which enabled banks to write off lost loans and the increase of their asset base because of rapid loan expansion that have been entirely responsible for the decline in the bad loan ratio.[6] There is no evidence that Chinese banks have begun to price their loans correctly. In fact to date there is hardly any differential pricing between banks. This year, the central bank has had to impose restrictions on banks from lending to 11 major industries which are deemed to be suffering from overcapacity or overheating. If banks adequately control or price their own risks, such restrictive measures would have been unnecessary. Therefore, nobody should be so na?ve as to believe that Chinese banks have stopped lending to bad credits or that they are now pricing their risks correctly. If the central bank does not believe it, why should you?
By having over-lent and under-priced their risks, Chinese banks have in effect substantially subsidized firms. All the money required to recapitalize Chinese banks can be viewed as bank subsidies to firms without which the collective profitability of Chinese firms would have been much less—if there were any profits at all. Chinese women are taller now than 25 years ago because most are wearing high heels. Chinese firms appear taller because they stand on Chinese banks.
At the Margin
The World Bank has made much of the fact that corporate profits have increased since the late 1990’s. Indeed, NBS data show that, assuming a 30% effective corporate income tax rate, net profits have increased at an annualized rate of 28% from 1999 to 2005. But one has to keep in mind that China’s fixed asset investment also has been increasing at about 25% to 30% per year during the same period. You would expect more profits on a $5 million investment than a $1 million one. Indeed the sheer number of firms in the database alone has increased by about 100,000 to 250,000 in 2005 from 150,000 in 1999. There is nothing surprising or remarkable about aggregate profits increasing in light of the massive investments and the number of new entries.
What about return on investment, however? The average return on equity for these firms has grown much more modestly. Assuming a 30% effective corporate income tax rate, ROE has grown at an annualized rate of 3.8% since 1999, and it did not increase between 2004 and 2005. In my September essay in the Far Eastern Economic Review, I attributed part of the increase to net productivity gains which have grown more than 5% per year during the same period of time. But that is not the full story. The picture is not complete without looking at the role of banks. To see where the profitability gain comes from, we need to go through some simple “margin analysis.”
Businesses look at different margins to understand different aspects of operating performance. Gross margin is calculated by dividing the difference between sales revenue and costs of goods sold by sales revenue. It is used to gauge whether the movement in input and output prices is favorable or unfavorable for the business. Gross margin for the industrial firms in the NBS database has fallen by 3.51% per year from 2000-05, indicating that these firms in general have been under severe pressure from either rising prices of raw materials or falling prices of finished products or both.
But net margin has risen modestly during the same period. Net margin is the ratio of net income divided by sales revenue. Assuming 30% effective corporate income tax rate, net margin for these firms has increased by 0.49% per year.
What explains the gain in net margin in spite of falling gross margin? We can find out the answer by examining the “EBIT margin,” which is the ratio of earnings before interests and taxes divided by sales revenue. This margin indicates how much a company has earned from a unit of sales before paying interests and taxes. The EBIT margin has fallen by 0.47% per year during this period of time. A falling EBIT margin indicates that companies have earned increasingly less before interest payments. Productivity gains obviously have not been sufficient to offset the decrease in gross margin. This is where banks, once again, come to the rescue. Clearly, interest rate cuts more than offset the decrease in EBIT margin, which is the only reason why the net margin increased.
Searching for elusive returns or for a cure?
By now, we know that average ROE for Chinese industrial firms is not high by international standards. We have also established that undistributed profits do not finance more than half of corporate investments. Neither do corporate savings. The increase in net margin for Chinese industrial firms in the past 7 years comes entirely from interest rate cuts. Banks have played the rich uncle, rescued and subsidized Chinese firms with huge amounts. So are Chinese banks profitable on average in comparison with their international peers? Not at all. The past three years were the best ever for Chinese banks. Yet, their average net return on assets, at 0.4%, is the lowest in Asia. Even without taking into account the need for provisions for bad loans, they are still the least profitable among their Asian peers with their pre-tax, pre-provision profit return on asset of merely 1.1%.[7] Search no further for China’s elusive superb returns. They do not exist.
Low return on capital is the hallmark of the growth model driven largely by fixed asset investments, such as China’s. This issue was laid to rest years ago when a number of leading economists, notably Professors Lawrence Lau, Alwyn Young and Paul Krugman, presented conclusive evidence that the so-called Asian miracle was produced much more by greater input than productivity gains.[8] They were proved right beyond the shadow of a doubt by later events, particularly the 1997-98 Asian financial crisis. China is no exception, only worse. Its growth has been particularly costly, wasteful and inefficient, pockets of excellence notwithstanding.
There is no lack of examples of sustained high growth in the annals of world development history. However, no economy in the history of mankind has ever come even close to investing as much as China does today to produce a double digit growth. I am not a big fan of using the incremental capital output ratio or ICOR to measure the efficiency of capital allocation and use for an economy because capital is not the only factor contributing to economic growth. But at times the ratio provides a reality check.[9] If it takes 25% of 2005 GDP to recapitalize Chinese banks, adjusting for such losses would shave 3.1% off the GDP growth number each year for the past 10 years, reducing the average growth rate to only 6%, as opposed to 9.1% per year.
I characterized China’s as “borrowed growth” in an article published three years ago.[10] My basic thesis then, as is today, is that China’s growth is enabled and sustained by its high savings rate and made possible by banks. But it is very costly and much of the cost is reflected in bad loans in the banking system. Sooner or later the country will have to pay for the losses in the banking system, as other Asian countries have done. It is in this sense that China’s growth is borrowed. The high savings rate is likely to decline sharply as the country ages precipitously in about 10 years time when baby-boomers retire due to China’s decades old birth control policies. To sustain economic growth, China needs banking reforms in order to improve the efficiency of how it allocates and prices its capital and other resources.
China has since then embarked on major banking reforms, and meaningful progress has been made. But Chinese banks are not out of the woods as yet. The economy’s growth continues to be driven by excessive liquidity, and so is costly and inefficient. Improving profitability, returns and efficiency remains the highest priority. While China is on the right track in her search for a cure, the last thing she needs is someone in a doctor’s white gown to come along to tell her she is in excellent health. Fortunately, the leadership knows better. The policy of the central bank to raise interest rates, mop up excess liquidity, curtail lending to overheated industries and generally increase the cost of capital is correct and necessary for sustained growth in the long term.
Weijian Shan is a partner of TPG Newbridge, a private equity firm. The author wishes to thank Forrest Chan for his first-rate research support.
[1] See “World Bank China Delusions,” by Weijian Shan, Far Eastern Economic Review, September 1, 2006, and “China’s growing dwarf,” by Wall Street Journal Asia, September 13, 2006.
[2] See “The great Chinese profits debate,” by Stephen S. Roach, Morgan Stanley Research, October 6, 2006.
[3] The data series published by National Bureau of Statistics include industrial firms with annual revenue of more than Yuan 5 million.
[4] See “World Bank China Delusions,” Weijian Shan, Far Eastern Economic Review, September 1, 2006.
[5] See “China”, Moody’s Investors Service, September 2006.
[6] “Reported NPLs fell in 2005,largely due to the government's carve-out of…legacy bad debt. Otherwise, NPL amounts rose a moderate 16.9% YoY at end-March 2006, while NPL ratios were largely flat as banking assets grew rapidly.” Page 2,“Banking System Outlook 2006: China,” Moody’s Investors Service, July 2006.
[7] For the analysis of relevant data, see “Banking system outlook: 2006,” Moody’s Investors Service, July 2006.
[8] A good summary of this work is in “The myth of Asia’s miracle” by Paul Krugman, Foreign Affairs, November/December, 1994.
[9] The ratio, which is calculated by dividing investment as a percentage of GDP by GDP growth rate, roughly measures how much investment is required for a unit increase in GDP.
[10] “China’s borrowed growth” by Weijian Shan, The Asian Wall Street Journal, September 3, 2003.
