by Michael Pettis
Without wanting to sound obsessive I want to re-emphasize the ideas of Minsky on balance sheets. (See my article last month.) In trying to judge the probability of a crisis we need to think not just about the probability and impact of positive or negative events in and of themselves. It is extremely important that we also understand the balance sheet mechanics that force the system into self-reinforcing behavior. The structure of the balance sheet itself, in other words, is as important in determining the impact of adverse events as the direct impact of those events on the asset side of the balance sheet.
This is, of course, one of Hyman Minsky’s key insights, and in that light I thought I would pass on part of a lecture by James Galbraith on Keynes which Galbraith gave at the 5th annual “Dijon” conference on Post Keynesian economics, delivered in Copenhagen in May last year. In the lecture Galbraith made a brief detour on the subject of Minsky in which he said:
Hyman Minsky developed an economics of financial instability, of instability bred by stability itself…Minsky’s approach, very different from Godley’s, is conceptual rather than statistical. A key virtue is that it puts finance at the center of economic analysis, analytically inseparable from what is sometimes called real economic activity, for the simple reason that capitalistic economies are run by banks. And, of course, his second great insight is into the dynamics of phase transitions: the famous movement from the hedge position to the speculative position to the intrinsically unsustainable, doomed to collapse ponzi position which arises from within the system and is subject actually to formalization in the endogenous instabilities of non-linear dynamical models.
To grasp what Minsky is about, it seems to me, is to go immediately beyond the coarse notion of the “Minsky moment,” a concept which implies falsely that there are also non-Minsky moments. It is to recognize that the financial system is both necessary and dangerous, that strict financial regulation is both indispensable and imperfect.
Any attempt to predict the likelihood and extent of a breakdown in an economic system – country, region, or company – that starts only from the asset/operational side of the economic entity (what Galbraith refers to above as real economic activity), without taking into account the feedback mechanisms inherent in the relationship between the asset and liability sides, is pretty useless.
What’s more, the recent history of disturbances in that economic entity tells us nothing about the future impact of similar disturbances – as long as the balance sheet structure is changing, and as Galbraith reminds us, the lack of instability during previous disturbances will itself change the structure of the balance sheet. Stability is itself destabilizing, as Minsky warned us, because it changes the nature of the relationship between the two sides of the balance sheet.
Commodities and growth
Volatility, in other words, is a function not just of volatility in real economic factors, but also, and perhaps even more so, of the structure of the balance sheet. The structure of the balance sheet can either smooth out normal economic fluctuations or it can turn them into highly destabilizing events.
One example of a destabilizing feedback mechanism worth pondering is the relationship between commodity prices and Chinese growth. Here is a very interesting article from last week’sFinancial Times:
Chang Zhenming, chairman of Citic Pacific, is unambiguous about the significance of his company’s Sino Iron mine in the desolate, red-soiled Pilbara region of Western Australia. “The whole of China is watching this project,” he says.
More to the point, China is watching with some trepidation as his Hong Kong-listed company faces increasing cost overruns and delays. The stakes are high. Mr Chang says Sino Iron is four times bigger than any iron ore project at home.
While outside observers often fear Chinese companies are unstoppable juggernauts in their ravenous pursuit of the world’s minerals, much of this perception is inaccurate. China’s international resource expansion is not running smoothly. The world’s second-biggest economy had hoped it would more easily control its economic destiny by taking huge mineral stakes, robbing companies such as BHP Billiton, Vale and Rio Tinto of the ability to dictate commodity prices.
But the Sino Iron project, far from being a showcase for China’s might, has become instead a cautionary tale of the difficulties Chinese enterprises face as they seek to expand abroad. When it was first conceived in 2006, the total cost was estimated at under $2bn. By now, it has already cost Citic Pacific $7.1bn. Analysts at Citigroup calculate the bill could swell to a possible $9.3bn, while others say they expect the ultimate bill will be closer to $10bn. The mine is at least two years behind schedule.
“This is no longer about commercial goals,” says a senior executive at one leading Asian trading company with extensive sourcing operations in Australia. “It is about Chinese machismo. They have plonked down too much money to pull out now.”
Leaving aside that rather interesting and even surprising last paragraph, one obvious comment on this article is that vastly overspending on a project of this nature is not at all surprising in a system in which privileged operators have near infinite access to cheap funding, little accountability, and no budget constraints for any project that can be proposed as being of national importance (and it is astonishing how many projects fit under that category). But the lesson I want to draw is a very different one – a balance sheet lesson.
In projects like this, and in the extent of commodity stockpiling we have seen more generally, China has taken a huge long position. Some analysts argue that China, by buying far more in the way of commodities and commodity producing companies than it requires for its immediate needs, is hedging its future demand.
Others make an even stronger case. Dambisa Moyo, a former investment banker turned economic writer, has argued in her book Winner Take All that the world is facing a crisis in the form of a commodity shortage. According to a recent review in the Guardian,
If Moyo’s calculations are correct, we are in big trouble – which makes the central premise of her book, Winner Takes All, all the more arresting. Governments across the world, she writes, have singularly failed to grasp what’s coming – with one sensational exception. “Simply put, the Chinese are on a global shopping spree.” State-sponsored Chinese corporations are busy buying up commodities across Africa, North America, the Middle East, South America – anywhere they can – in a concerted strategy to seize control of resources before the rest of the world wakes up to the looming crisis.
They’re striking deals with what she calls the “axis of the unloved” – developing countries rich in commodities but poor in political and economic capital – in return for much needed investment, employment and infrastructure. Extravagant shoppers, the Chinese are happy to pay over the odds, treating their trading partners not as poverty-ridden charity cases nor political pariahs but valued commercial equals.
But when the resources begin to run dry, the consequences will be catastrophic. Already, since 1990 at least 18 violent conflicts worldwide have been triggered by competition for resources. If nothing is done now, warns Moyo, commodity wars on a terrifying scale are all but inevitable.
Inverted balance sheets
Perhaps it is my natural skepticism, but we have heard warnings like these many times before, and they have usually proven to be spectacularly wrong largely because they are based on projections of recent trends that are clearly unsustainable. In my opinion the next few years are not going to see soaring commodity prices but rather collapsing commodity prices, in large part because it has been China’s unsustainable investment boom that has both driven demand up ferociously (accounting for only 10% of global GDP China nonetheless absorbs roughly 40% of global copper production and nearly 60% of global iron ore and cement production) and driven up investment in extractive industries.
Once China brings down its infrastructure investment rate, the combination of declining demand (in fact China has stockpiled so much that it will soon turn from importing copper, iron ore, and other commodities to exporting them) and expanding supply is likely to have a very deleterious effect on prices. In my opinion China is paying overly high prices in a market in which prices are likely to drop sharply.
But reasonable people can differ on whether or not commodity prices are going to rise substantially. What reasonable can never do is place too much confidence in their predictions. Dambisa Moyo may be right that commodity prices will soar, and remain permanently high. I doubt it, but the real reason I think China is making a mistake in stockpiling commodities is not because I think prices will inevitably decline, but rather because it is a risky balance sheet strategy for China. It exacerbates the volatility impact of commodity prices, which are already very volatile, and this brings us back full circle to Hyman Minsky.
Why is stockpiling a bad strategy for China? It is risky because of the inverted relationship between Chinese growth and commodity prices. It is widely agreed in the commodity industry that the biggest cause of rising commodity prices in the past decade has been the ferocious growth in Chinese demand, and this growth has been primarily a consequence of Chinese investment growth. If China keeps growing rapidly, of course, we may very well see higher commodity prices in the future, but – and this is the problem – if China slows significantly, the price of commodities is likely to decline, at least in the next few years.
So China has effectively made a big bet on commodity prices, and it “wins” the bet if it continues to grow quickly. It “loses” the bet, however, if its growth rate slows sharply. This is what I referred to as an “inverted” capital structure in my 2002 book, The Volatility Machine. An inverted structure is the opposite of a hedged structure – when the asset/operational side of your balance sheet does well, your liability side also does well, but when the asset/operational side does badly, the liability side does too.
Inverted balance sheets exacerbate volatility – good times are automatically better than they otherwise would have been and bad times are automatically worse. Countries (or companies) with inverted balance sheets are more volatile than countries with hedged balance sheets, and unless you can get all your speculative bets right, this higher volatility lowers growth over the long term. Inverted balance sheets, I argued in my book, are one of the key differences between countries that are able to recover successfully from crisis and countries that aren’t, and I would propose that this may be one of the differences between countries that can escape the middle income trap and countries that can’t.
Of course a country’s balance sheet is affected by a lot more than just commodity stockpiling. There are many other aspects of China’s balance sheet that matter, but I would argue that good liability management consists of eliminating sources of volatility in the balance sheet by structuring it in ways that cause the performance of the liability side and the asset side (or, to put it another way, the structure of expenses and the structure of revenues) to move in opposite ways, not in the same way.
This isn’t happening – in at least one aspect of the national balance sheet, commodity stockpiling. To take another example, hot money flows are automatically volatility enhancers – when the economy is growing quickly, money pours into the country and causes even more growth, but when the economy gets into any trouble, money flees and so causes even more contraction.
China in principle has capital controls, which should prevent this from happening, but in practice Chinese capital controls are extremely porous, and as Chinese prospects have gotten worse in the past two years, we have seen what is clearly a surge in capital flight. If China is serious about internationalizing the renminbi and relaxing capital controls it will only increase the balance sheet inversion (which is why I think we are going to see a reversal of RMB internationalization in the next few years).
Or to take two more obvious examples, first, asset based lending – for example against real estate – is also a source of balance sheet inversion. When asset prices rise, the value of debt collateralizing the assets also rises, but when asset prices drop the debt becomes less credible and its implicit cost to the economy rises. Second, borrowing short term, or borrowing in a foreign currency, has the same risk profile. When the country is doing well, the real cost of short-term or foreign currency debt declines, only to surge when the economy gets into trouble.
Sometimes inverted capital structures are inevitable, but liability management consists, in my opinion, of identifying ways of eliminating inversion when you can and embedding as much hedged liability structures as you can, so as to make the overall economy less, not more, volatile. In the case of China, stockpiling commodities is exactly the wrong thing to do – but of course it is hard to convince anyone that this is the case when we are in the “good” part of the volatility cycle.
My baby drove off in my brand new Cadillac
It is only when conditions turn for the worse that everyone recognizes – albeit usually too late – the risk. We see this happening in Europe. When Europe was booming and the borrowing costs for the peripheral countries were converging with that of countries like Germany, it was hard to convince anyone that this was an extremely risky balance sheet structure.
Now that Europe is in crisis and the very source of interest rate convergence – the euro – is causing a massive divergence in borrowing costs, everyone recognizes, albeit too late, the danger of highly inverted balance sheets. But, as I pointed out in my book, no matter how often history repeats, during the good part of the volatility cycle it is brutally difficult to convince anyone of the need to change the structure of the balance sheet. The riskier and more inverted it is, the more money everyone makes. All you can really do is write about it, and point out the occasional country – like Chile in the past two decades – that have learned, however temporarily, how the volatility machine embedded in balances sheets works. Ed Chancellor wrote about this process recently for the Financial Times.
To point out a slightly lighter story of balance sheet inversion here is another Financial Times article that I found very interesting:
Cash-strapped local governments in China have begun auctioning off fleets of officials’ luxury cars as part of efforts to bolster revenues hit by the country’s slowdown. Wenzhou, a south-eastern coastal city hit hard by the cooling economy, sold 215 cars at the weekend, fetching Rmb10.6m ($1.7m). It plans to sell 1,300 vehicles – 80 per cent of the municipal fleet – by the end of the year.
Government revenues from tax and land sales in Wenzhou have been declining after years of heady growth. With the city’s risk-taking businesses struggling to pay back debts, the burden has fallen on the local government to turn things around. State media noted the auctions would directly boost the city’s coffers.
Wenzhou is not alone. Across the country, from Kunming in the south to Datong in the north, officials have been tightening their belts, paring back on banquets, curtailing travel and trimming the fleets of tinted-window luxury cars that have long been standard issue – even in the middle ranks of government.
Buying fleets of expensive cars when everyone else is buying them, when the economy is booming, and selling them when everyone else is selling them, when times get tough, is a great way to lose money just when you can least afford it. This in itself is not a serious balance sheet problem for China, since the municipal losses are gains for people who want to buy luxury cars on the cheap, but this story is interesting for two reasons.
First, it shows how quickly perceptions have changed. Just a few years ago it seemed so inconceivable that we would face tough times that no one really questioned the wisdom of extravagant spending, but now clearly those questions don’t seem so absurd.
What is especial worrying about this story is not just that municipalities have splurged on cars in recent years. They have also see a surge in personnel, and larger than ever numbers of workers depend on the solvency of municipal governments for their paychecks. This solvency is becoming a big issue, and unfortunately the only way to solve it (temporarily, of course) seems to be by igniting another property bubble.
Second, I can’t help but see this except as a part of a bigger process of wealth transfers from municipalities (and so households in general) toward the buyers of distressed assets, who tend already to be quite wealthy. My guess is that for anyone with lots of liquidity and no real hurry to invest it, the next few years are going to produce quite a lot of bargains at the expense of the poor and middle classes, who will inevitably foot the bill.
In itself this story is more amusing than serious, but it does illustrate, I think, the way certain types of systems create incentives that tend to exacerbate volatility, and a thorough analysis of the risks associated with a country like China requires an understanding of the incentive structure and how it builds up balance sheet inversions. To continue on this topic, and at the risk of making this issue of the newsletter look like an advertisement for the Financial Times, let me turn to one last FT article, this time from the FT/Alphaville section, which has a habit of teasing out very interesting stories long before they are widely noticed.
According to an entry earlier last week:
ChinaScope reports that China’s total outstanding foreign debt was $751.26bn at the end of March 2012, according to data released Monday by the State Administration of Foreign Exchange (SAFE).
Here’s the trend to date, also courtesy of ChinaScope. As we can see not only is the foreign debt getting larger (in particular the SAFE portion) it’s getting shorter-term in duration too.
We are far from having in China a risky external debt structure, but this does bring up two issues. First, rising external debt simply adds to the many ways in which the national balance sheet has built up instability. This often happens in the late stages of an unsustainable credit boom because as stresses in the system appear, they are often resolved by structures that are, by my definition, inverted. As Chinese companies find it harder to borrow in RMB, for example, they increasingly take to dollar borrowing.
And as Chinese companies find it harder to borrow long-term, they borrow more short term. As the price of their commodity stockpiles declines, they add to their hoard to reduce average prices. As perceptions of financial fragility rise, the system switches even more to collateralized borrowing. We don’t know what the cumulative impact of all this balance sheet inversion is, but we need to acknowledge that the range of expected outcomes has become more volatile.
The second issue is just a history reminder. When Brazil went through its own debt financed investment boom in the 1960s and early 1970s, during the period of the Brazilian “miracle”, most of it was domestically financed. By the mid-1970s, however, Brazil began reaching domestic debt capacity limits, and so the economy began slowing.
The party, however, didn’t quite end. At around the same time the huge increases in oil prices had created massive petrodollar surpluses that weighed on bank balance sheets, and banks were eager to lend them out. Fortunately for them (or unfortunately, as it turned out), the developing countries, including Brazil, were able to turn to the banks and borrow their way through the economic slowdown of the mid-1970s.
The rest, of course, is history. Countries like Brazil were able to continue overinvesting, and continued growing in the late 1970 even as the US and Europe slowed (sparking much excited talk of “decoupling”). This went on until debt levels became unsustainable, and in 1981-82 credit abruptly stopped flowing. This was when the 1980s LDC debt crisis began.
I am not suggesting that China today is undergoing the same process as Brazil and that it will switch from domestic to external financing as the Chinese banking system finds it increasingly difficult to keep credit growth high. I certainly hope that this doesn’t happen, since it will simply allow China to postpone the necessary adjustment in its growth model for a few more years, but at the cost of a much more difficult adjustment. Brazil in the 1980s showed how painful that can be.
About the Author
Michael Pettis is a Senior Associate at the Carnegie Endowment for International Peace and a finance professor at Peking University’s Guanghua School of Management, where he specializes in Chinese financial markets. He has taught, from 2002 to 2004, at Tsinghua University’s School of Economics and Management and, from 1992 to 2001, at Columbia University’s Graduate School of Business. He is also Chief Strategist at Shenyin Wanguo Securities (HK). Pettis has an impressive work history on Wall Street, Latin America, Europe and Asia (see his blog China Financial Markets for a complete bio).