On building debt

By Michael Pettis

Before starting on the subject of debt I wanted to make a quick reference to something sent to me by Charles Horner, a senior fellow at the Hudson Institute.  I am glad to say that the overinvestment thesis is much more widely acknowledged today than it was even two or three years ago, but one myth, I think, is that most of the overinvestment excesses in China are concentrated in the real estate sector.  I have always argued that it is infrastructure where the most amount of investment has been wasted.

Its impossible to prove one way or the other, but Horner sent me a paper in the Oxford Review of Economic Policy, by Oxford’s Bent Flyvbjerg, with the rather alarming title “Survival of the unfittest: why the worst infrastructure gets built—and what we can do about it”, which suggests why we need to be so worried about infrastructure spending in China – aide from the fact that the numbers are simply huge.

In the paper Flyvbjerg looks at infrastructure projects in a number of countries (not in China, though, because he needed decent data) and shows how the benefits of these projects are systematically overstated and the costs systematically understated.  More important, he shows how these terrible results are simply the expected outcomes of the way infrastructure projects are typically designed and implemented.

It is not a very happy paper in general, but I am pretty sure that many people who read it probably had a thought similar to mine: if infrastructure spending can be so seriously mismanaged in relatively transparent systems with greater political accountability, what might happen in a country with a huge infrastructure boom stretching over decades, much less transparency, and very little political accountability?  Isn’t the potential for waste vast?

Who knows, but it seems that Beijing is increasingly worried about that possibility.  Here is an article from this week’s Caijing:

A golden but brief era for urban railway suppliers, builders and related companies across China appears to have ended in recent months.  Local governments nationwide have slashed infrastructure spending since last summer, and the urban rail business has slowed to a crawl after several years of rapid growth.

…NDRC imposed extremely tight regulations on the approval process for subway construction before 2008, said Jin Yongxiang, general manager of Beijing’s Dayue Consulting Firm, which advises subway projects.  Jin said “the situation took a 180 degree turn” in April 2009, when the State Council reduced the minimum capital requirement for urban rail projects to 25 percent. At that time, a NDRC source told Caixin, credit was loose and bank loans were easy to obtain.

“In some cases, the NDRC gave a nod to an urban rail project even if a local government had yet to meet the minimum capital requirement,” the source said. “With NDRC’s approval for subway projects, banks were willing to lend and would not hold city governments to the capital requirement.”  In some cities, though, enthusiasm for the urban railway building went too far.

For example, rail lines were built where few people live on the outskirts of the Hunan Province city of Changsha, said Wang Chengli, an urban transit professor at the city’s Central South University. Today, exit gates for some of the city’s finished subway stations lead to farm fields.  Wang said Changsha authorities installed far fewer kilometers of track in the city’s center than in its suburbs. Each project was approved by the central government, he added.

Zhang says China learned important lessons from the fast-track subway program. For example, he now thinks subways should never have been built in “many cities.”  “The only cities that should have built subways are super-large ones such as Beijing, Shanghai, Tianjin, Shenzhen, Wuhan, Nanjing and Guangzhou,” Zhang said. “Provincial capitals such as Shenyang and Taiyuan can handle their transit needs with a single, light-rail line.”  Subways can be uneconomical in smaller cities. Zhang said final costs for many projects were often much higher than a local government’s estimate.

Little by little the claim repeated by so many China bulls – that you can never spend too much on infrastructure – is being eroded. It is possible, it turns out, to waste a lot of money even on infrastructure, and if debt-fueled investment is being wasted in China, as I have been arguing for over half a decade, then without doubt debt must be rising at an unsustainable pace. Last week Bloomberg had this article, which suggests that even the official numbers, which show debt soaring, may be understating the reality:

Debt accumulated by companies financing local governments such as Tianjin…is rising, a survey of Chinese-language bond prospectuses issued this year indicates. It also suggests the total owed by all such entities likely dwarfs the count by China’s national auditor and figures disclosed by banks.  Bloomberg News tallied the debt disclosed by all 231 local government financing companies that sold bonds, notes or commercial paper through Dec. 10 this year. The total amounted to 3.96 trillion yuan ($622 billion), mostly in bank loans, more than the current size of the European bailout fund.

There are 6,576 of such entities across China, according to a June count by the National Audit Office, which put their total debt at 4.97 trillion yuan. That means the 231 borrowers studied by Bloomberg have alone amassed more than three-quarters of the overall debt.

The fact so few of the companies have accumulated that much debt suggests a bigger problem, says Fraser Howie, the Singapore-based managing director of CLSA Asia-Pacific Markets who has written two books on China’s financial system. “You should be more worried than you think,” he said of Bloomberg’s findings. “Certainly more worried than the banks will tell you.

Why debt matters

There is more to the article.  For example Huang Jifa, in the investment banking division at ICBC, reportedly says that local government loans aren’t a problem because the projects will generate returns, even if not immediately. “The money that Chinese local governments have borrowed is not like the money people borrowed in Europe or Greece,” Huang said in a Nov. 24 interview. “The Chinese government’s borrowed money is all invested. Many projects will have returns.”

Maybe, but I am pretty skeptical.  The problem of course is that it doesn’t matter that many infrastructure projects in China have returns.  I am sure many do (as did many projects in Greece, no doubt).  What really matters is whether all the various projects in the aggregate are generating greater returns than the debt servicing cost, adjusting of course for all hidden and explicit subsidies.  If not, then debt levels must be rising faster than the economy’s ability to service the debt.

Here, by the way, is another interesting related article, from last week’s South China Morning Post.

Auditors have uncovered 530 billion yuan (US$84.21 billion) worth of irregularities with local government debt, the National Audit Office said on Wednesday.  An audit report, published on China’s central government website, reveals some of the problems investment analysts had believed to lay beneath the 10.7 trillion mountain of debt that local governments had amassed by the end of last year.

The report, conducted for the last year budget year, found problems including 46.5 billion yuan worth of “irregular credit guarantees”, 73.2 billion yuan worth of loans secured against irregular collateral, 35.1 billion yuan spent on stocks, houses and polluting plants and 132 billion yuan worth of expenditure not made by its approved deadline.

“A fifth problem is the fraudulent and underpayment of registered capital in financing vehicles, which amounted to 244.15 billion yuan,” the report said.  The local governments involved have been ordered to correct wrongdoings, but the clean-up work remains less than half done in some areas, the report shows.

Xinhua later added an update to the story in which they focused on the amount of the irregularity that had been recouped:

Nearly half of the misused funds uncovered in the auditing of China’s local government debts in 2010 has been recouped, authorities said on Wednesday.  Of the 530.9 billion yuan (about 84.3 billion U.S. dollars) of misused funds uncovered for the year 2010, around 259.2 billion yuan had been recouped by Oct. 2011, the National Audit Office (NAO) said in a report on the year’s auditing progress.

…Violations of the management of local debts involved illegal guarantees for local debts, misdirected funds to capital, property and energy-consuming markets, and the operation of fake investment companies, the report said, adding that the governments have been moving actively to correct the irregularities.

This particular story is less important than the frequency with which we hear similar stories.  Every credit bubble in history seems to have been accompanied by a surge in accounting “irregularities”.  Irving Fisher explained why in the 1930s, and Hyman Minsky also indirectly showed why this might happen.  It is, I guess, a necessary accompaniment to out-of-control credit expansion.

I get many calls from investors and journalists on these debt-related topics, usually because they worry about the ability of specific borrowers and sectors to service the reported debt, and I always make the same response.  The debt will be serviced.  One way or the other it will be assumed by the central government through the banking system.

But this is not the important issue.  The important issue is that it is clearly proving impossible to keep GDP growth levels high without explosive debt growth, and there are serious debt capacity limits to this kind of debt growth.  I have no idea where the debt will next show up, or what the next debt panic will be (I suspect this year it will be SOE debt), but I have no doubt that there will be more of these debt panics.  This is not an accident.  It is intrinsic to the way the development model works.

The problem, then, is not that there will be defaults.  The problem is that the only alternative to default is to service the debt, and this is what will cause the real damage to the economy.  If the economic benefits generated by the investment are less than the correctly-valued debt-servicing costs, as they almost certainly are, the difference has to be made up in the form of a transfer of resources from some sector of the economy.

As we saw in the last debt crisis, a decade ago, debt-servicing costs are only manageable in China thanks to financial repression – i.e. extremely low lending rates funded by even lower deposit rates — which implies a huge transfer, equal to several percentage points of GDP annually, from household savers to corporate and government borrowers.  Households, in other words, typically clean up banking messes.

Only consume!

The problem with this solution is in what it implies about future growth in demand.  If investment is being wasted, it must be reduced or it will create a debt crisis eventually. If the external environment is tough, the demand impact of a sharp drop in investment cannot be made up for by a surge in the trade surplus – in fact the trade surplus may actually contribute negative demand.  So where will the demand come from needed to pull the Chinese economy? The only possibility is a surge in domestic consumption.

Can consumption possibly surge?  No, not if the household sector is going to be forced to clean up the banking mess again.  This is the same problem that caused household consumption to drop after the last banking crisis from a very low 46% of GDP in 2000 to an astonishing 34% in 2010.

In that light, it was interesting to see this article in Wednesday’s South China Morning Post:

The Chinese government is planning new policies to boost domestic consumption, especially of vehicles and appliances, in a bid to offset the effects of sagging export demand, the China Daily reported on Wednesday, quoting a government official.  With tax rebates on vehicles and domestic appliances either having expired or due to expire, the government is working on new measures, said Huang Hai, former assistant minister of commerce and a member of the economic and trade policy consulting committee linked to the Ministry of Commerce.

These may include subsidies for families living in affordable housing that buy electrical appliances and for consumers planning to change cars, the paper said.  The newspaper also quoted a Ministry of Commerce spokesman as saying that the ministry was considering new programmes to expand consumption, with details to be announced next week.

Huang also said over 10 government agencies, including the Ministry of Commerce, the National Development and Reform Commission (NDRC) and the Ministry of Finance, are expected to co-operate and propose concrete plans to boost consumption at a meeting slated for April.

“If at first you don’t succeed, try, try again,” WC Fields advised but, he added, “then quit. There’s no point in being a damn fool about it.” I don’t think this new attempt to boost consumption has any chance of succeeding, any more than similar polices did in 2009 and 2010.

Sure, consumption of automobiles and white goods surged back then, just as you would have expected given the subsidies, and they will surge again no doubt, but since those subsidies were ultimately paid for by the household sector, the policies did not translate into an overall surge in consumption because there was no net increase in household wealth. In fact during both of those years consumption continued to decline sharply as a share of GDP.

The Financial Times version of this story makes a classic mistake:

China has ample room to stimulate consumption. Household spending accounted for half of gross domestic product two decades ago but dwindled to just 33.8 per cent of GDP in 2010, a record low for a major economy in peacetime. China is probably now at a turning point in that consumption is beginning to become a bigger force in the economy, but this will be a “longer-term process”, said Zhu Haibin, an economist with JPMorgan.

For years China bulls have been arguing that because the Chinese save so extraordinarily much money, there is plenty of room to stimulate consumption – just get them to save a little less.  The problem with this reasoning is that consumption is not low because Chinese households save a lot (they save in line with other Asian countries as a share of their income, and less than some).  It is low because household income is such a low share of GDP.

It isn’t about household savings

The only way to boost household consumption is either to redistribute income from the low-consuming rich to the high consuming poor, or, better yet, to redistribute wealth from the state to households.  Both of these have serious political implications that have to be resolved and are unlikely even to be addressed with consumption subsidies.  After five years of this argument, during which time consumption has plummeted relative to total savings, you would think they would start to abandon the idea that all we need to do to get consumption to surge is to reduce household savings a little.

When we add in the possibility of a continued decline in house prices throughout China, we may start to feel some kind of wealth effect dragging consumption growth down even further as a share of GDP, although I am not sure I am too worried about that.  The housing boom seems to have mainly benefitted speculators, and I don’t think that it translated into a significant increase in consumption when housing pieces were on their way up.  In that case it shouldn’t matter too much on the way down either, although it is better to wait and see what happens.

Nonetheless on Tuesday Xinhua has this to say about housing prices:

Sales of both new and existing homes in Beijing plummeted in 2011 as a result of the government’s efforts to cool down the runaway property market.  New home sales in Beijing dropped 18.4 percent to 90,605 units in 2011 from a year ago, the Beijing News reported Monday, citing data from the city’s housing regulator.

In terms of square footage, sales slumped 22.4 percent to 9.56 million square meters last year, falling below 10 million square meters for the first time in six years, the paper said.  Existing property transactions also plunged. Sales in 2011 shrank 38.2 percent to 121,512 units, hitting a three-year low.  Monthly sales of existing homes have stayed below 10,000 units since April, the paper said.

Increased down-payment requirements and mortgage rates, as well as limits on home purchases, led to the declines, the paper said, citing Zhang Dawei, a chief analyst with Centaline Property.  Consumers will likely expect further price drops in 2012, as the government has reiterated that it will maintain the policies, Zhang said.  Property prices in Beijing are likely to fall 10 percent to 20 percent over the next six to 12 months, he said.

On a related note, on Tuesday, Li Yan, a young musician I know, called me up to ask me for some advice.  His parents, who come from and still live in a small town in Hebei province, wanted to buy him an apartment in Beijing, where he lives and works, because they were sure that in a few years the same apartment would cost an awful lot more. He wanted to know from me whether it made sense to do so. He had heard prices are coming down rapidly and asked me whether he should tell his parents to wait. Yes, I told him, wait.

I mention this story because Li Yan is a 21-year-old kid who has just started to become famous among Chinese youth for his wild antics and wilder music, and I suspect he thinks about and discusses finance and real estate as often as I think about and discuss life on Venus. Yet even he has heard that real estate prices are dropping.  Two years ago, of course, no one in China had any doubt that real estate prices can only go up.

This is an abbreviated version of the newsletter that went out two weeks ago.  Academics, journalists, and government and NGO officials who want to subscribe to the newsletter should write to me at [email protected], stating your affiliation, please.  Investors who want to buy a subscription should write to me, also at that address.