Will China have to choose between social stability and long-term growth?

Wow!  There are now rumors that Chinese net credit growth in January was substantially higher than the already-astonishing rumors of RMB 1.2 trillion I reported last week.  I will get to that at the end of this entry, but I wanted first to discuss a possibly important issue related to credit intervention.

It is probably not at all controversial to suggest that the way governments in the US, China and elsewhere respond to the current crisis will determine economic growth prospects for the next decade and more, but it is probably also worth repeating this point as often as possible.  In the panic to respond swiftly to some of the short-term problems facing policymakers, it would be easy for them sometimes to forget the longer-term impact of current policy responses, and so saddle us for many years with unwanted consequences.

Over the weekend I was reading a paper by Gonzalo Fernández de Córdoba (Universidad de Salamanca) and Timothy J. Kehoe (University of Minnesota), called “The Current Financial Crisis: What Should We Learn from the Great Depressions of the Twentieth Century?” Basing their work on Great Depressions of the Twentieth Century, published in 2007 by the Federal Reserve Bank of Minneapolis, in which Timothy Kehoe and Edward Prescott, together with a team of 24 economists from around the world, analyze a number of “great depressions” experienced by various countries in the 20th Century, they try to determine the impact of policy on the subsequent severity of the contraction.

Although I always worry about ideological predispositions in these kinds of analyses (one group of economists always seems to find that government intervention made things worse, while another always seems to find that in fact specific policies helped), some of the examples they use – in Latin America primarily – involve countries and histories with which I am pretty familiar, and at least this part of their analysis rings true to me.

Based on the data analyzed in the book, they conclude that massive public interventions in the economy to maintain employment and investment during a financial crisis will, if they distort incentives enough, make things much worse.  I guess that probably wouldn’t come as a very controversial statement to anyone, but what interested me was that they seemed to focus especially on ways that governments have intervened in credit markets and in investment decisions.  Two examples were especially illuminating, Mexico and Chile – which both experienced massive crises beginning in 1982, the year which usually signals the beginning of the LDC Debt Crisis (or the “lost decade”, as Latin Americans call it).  Their policy responses in the financial sector were radically different:

In 1982 in Chile, banks that held half of the deposits were suffering severe liquidity crises.  The government took control of these banks.  Within three years, the Chilean government had liquidated the insolvent banks and reprivatized the solvent banks.  The government set up a new regulatory scheme to avoid mismanagement.  These new regulations allowed the market to determine interest rates and the allocation of credit to firms. The short-term costs of the crisis and the reform in Chile were severe, and real GDP fell sharply in 1982 and 1983.  By 1984, however, the Chilean economy started to grow, and Chile has been the fastest-growing country in Latin America since then.

In 1982 in Mexico, the government nationalized the entire banking system, and banks were only reprivatized in the early 1990s.  Throughout the 1980s, in an effort to maintain employment and investment, the government-controlled banks provided credit at below-market interest rates to some large firms and no credit to others.  Even the privatization of banks in the early 1990s and the reforms following the 1995 crisis have not been effective in producing a banking system that provides substantial credit at market interest rates to firms in Mexico.  The result has been an economic disaster for Mexico:  Between 1982 and 1995, Mexico experienced no economic growth and has grown only modestly since then.

The differences in economic performance in Chile and Mexico since the early 1980s have not been in employment and investment, but in productivity.  In Chile, unproductive firms have died and new firms have been born and grown.  Workers and capital have been channeled from unproductive to productive firms.  In Mexico, a poorly functioning financial system has impeded this process.

GDP per working age person in Mexico declined substantially in the 1980s and only began recovering by 1988, but a second banking crisis in 1995 eroded much of the recovery and as of today it has still not reached its 1982 peak.  In Chile, the decline at first was much sharper.  In two years GPP per worker in Chile dropped by around 20%, which it took six years to happen in Mexico.  However productivity growth surged thereafter so that by 1988 it had fully recovered to 1982 levels and as of today it has doubled.  Chile, as most of us know, has been for the past twenty years the fastest growing country in Latin America, even though it as among the worst hit by the debt crisis.

The main point the paper seems to want to make is that intervention in the allocation of credit had a huge impact on the way the country was able (or not) to recover from the crisis and regain productivity growth:

Japan suffered a financial crisis in the early 1990s and followed similar sorts of policies as Mexico, keeping otherwise insolvent banks running, providing credit to some firms and not others, and using massive fiscal stimulus programs to maintain employment and investment. Japan has stagnated since then.  Finland also suffered a financial crisis in the early 1990s and followed similar sorts of policies as Chile, paying the costs of reform and letting the market dictate the allocation of credit to the private sector. The Finnish economy has grown spectacularly since then.

What implications this might have for Chinese policy-making in response to the current crisis?  Again, we always need to protect ourselves from conclusions that owe more to ideology than evidence, but at the very least we should consider the possibility that massive intervention in the banking system, for all the short-term countercyclical benefits (i.e. banks are forced to expand, to satisfy policy interests, rather than contract, to satisfy commercial interests) can create serious enough distortions that Chinese growth for the next decade or so might be sharply constrained.  In their words:

We need to avoid implementing policies that stifle productivity by providing bad incentives to the private sector.  With banks and other financial institutions in crisis, the government needs to focus on providing liquidity so that banks can provide credit at market interest rates, and using the market mechanism, to productive firms.  Unproductive firms need to die.  This is as true for the automobile industry as it is for the banking system. Bailouts and other financial efforts to keep unproductive firms in operation depress productivity.  These firms absorb labor and capital that are better used by productive firms.  The market makes better decisions than does the government on which firms should survive and which should die.

Of course someone will inevitably argue that it actually makes commercial sense for Chinese banks to expand loans now, since there is likely to be an implicit, or even explicit, guarantee that makes most new lending essentially risk-free.  Yes, of course, but that doesn’t change the underlying logic.  Banks will be channeling capital to companies not based on their economic prospects but rather based on the guarantee, and so little commercial distinction will be made between healthy and unhealthy borrowers.  My guess, and not a particularly controversial one I suppose, is that the provision of implicit or explicit government guarantees will have more to do with a company’s impact on employment than its economic prospects.

I don’t want to overstate the relevance of market versus government allocation of credit, but by the late mid-1980s, when I first started trading Latin American debt, it was pretty clear that Chilean banks were in much better shape than were Mexican banks, and were much more independent (Mexican banks were not privatized until the early 1990s). I specialized primarily in Mexican debt and bonds until I ended up running the Latin American trading desk, so losing my country focus, but it did always seem to me that the Mexican financial system was a lot less prudent than the Chilean, and government “guidance” had a very big impact on credit allocation.

Before someone suggests that perhaps poor guidance leading to credit misallocation might be less of a problem in well-governed China than in poorly-governed Mexico, I would argue that much of China’s recent growth came about because of the massive expansion in credit, and while the sheer size of the expansion guaranteed that there would be many years of bubble-like growth, we will only now, over the next three to five years, discover whether or not the capital was indeed misallocated on a massive scale.  I think it was.

The paper makes a point of saying that the difference in subsequent GDP growth between countries that intervened heavily in credit allocation versus countries that didn’t was not a function of different levels of employment, but rather different growth rates in worker productivity.  There were no noticeable differences in employment levels between countries that followed one strategy versus the other.

In that case one can make the argument that if the goal of policy is to minimize social disruption, the “Mexican model” may actually be better than the “Chilean model” because while neither model created a noticeable difference in employment levels, in Mexico an economic contraction roughly similar in magnitude took six years, versus the two years it took in Chile.  Mexico may have achieved this socially less disruptive adjustment at the expense of sharply lower levels of productivity growth over the long term, and perhaps this is the tradeoff that governments face in dealing with crises.  Japan, it seems to me, also chose a socially less disruptive model, in exchange for a lost decade of growth.

In other words the best policy advice for the government to maximize China’s growth prospects, based on the Fernandez and Hehoe paper, is probably politically unpalatable.  It would involve acknowledging that too much capital was allocated to production, and that a period of consolidation is necessary.  Unfortunately this consolidation means that capital migrate in a major way from less productive users to more productive users, which is just a bloodless way of saying that a lot of companies are going to have to be allowed to fail, and banks and financial markets should be weaned away from political control and encouraged to make their own commercial decisions.

But should this happen in the midst of a global crisis?  On the one hand, in China – and probably most other countries – real reform only seems to occur after a crisis, and so this is an important opportunity to get things right.  On the other hand global conditions are too ugly for China to allow bankruptcies to take their swiftest course, and so undermining the social pact, so a strong case can be made for intervening heavily now and reforming later.  Ultimately this is a political question that the Chinese must make: is there a tradeoff between long-term growth and short-term instability, and if so, which should China choose?

As I noted at the beginning of this entry, hot off the press is some related news about credit intervention.  In an entry last week I mentioned the astonishing RMB1.2 trillion increase in loans that had been unofficially reported for January.  This was a full 50% more than the previous monthly record, and nearly one-quarter of the total increase in 2008 (to be fair however January is traditionally always a big month for new lending).

Although this was seen widely as good news for the economy, since credit expansion will probably goose up the short-term GDP and employment numbers, I of course worried about exactly how much of this was real and, more importantly, how much of this will end up as future NPLs.  It seemed to me that even the most prudent and commercial banking system in the world cannot expand at this rate without shoveling in an awful lot of garbage, and loan expansion of this base represented a gamble on the duration of the global contraction.

Well, it seems I was wrong.  Reuters has just announced that net new lending may have actually been and even more surprising RMB 1.6 trillion – twice the previous monthly record and an amazing one-third of credit growth in all of 2008.  We will know by February 15 at the latest, when the PBoC publishes lending data, but if this is true (and the report was seen as highly credible by one of my friends at Reuters) it will probably goose the stock market up further while making people like me more worried then ever.  Since for me much of the Chinese growth explosion of the past several years was caused by a badly allocated credit boom, the idea that the solution to a slowdown is to jack up the credit boom even further is very worrying.  It is a little like the idea that the best way for the US to adjust to the decline in its debt-fueled household consumption binge is to replace it with a debt-fueled government consumption binge, although perhaps the US and China would choose very differently in the possible trade-off between long-term growth and short-term social stability.

At any rate if this number is true, and if these credit growth levels persist, at least it suggests China is very serious about contributing its share of global fiscal expansion.  This should be part of China’s negotiations with the US on trade relationships.

4 Responses to "Will China have to choose between social stability and long-term growth?"

  1. jye   February 9, 2009 at 1:59 pm

    Michael Pettis:I would argue that much of China’s recent growth came about because of the massive expansion in credit, and while the sheer size of the expansion guaranteed that there would be many years of bubble-like growth, we will only now, over the next three to five years, discover whether or not the capital was indeed misallocated on a massive scale. I think it was.My comments:”misallocated on a massive scale” against what measurement? If you use the performance of Wall street as benchmark, you don’t have to wait for “three to five years” to find that out. Sorry sir, you are proved to be wrong for now.

  2. Victor   February 9, 2009 at 6:59 pm

    Nice comments. Hey, have you seen the CICC note suggesting a relationship between sex ratio (male-female birth ratio) and savings rate. Apparently, it’s now 1.2, and within China, there is a positive relationship between sex ratio and savings rate. That is, the more boy children, the more likely parents are to save in order to ensure a bride for their son. Is that the source of global imbalance?

  3. Guest   February 10, 2009 at 3:21 am

    It would be interesting to see in what sector the loans have mostly been directed towards. Probably mostly SOEs I’d guess, since they’re the easiest channel.If it’s export oriented firms that are reaping most of these guaranteed loans, however, then it’s just going to make the beggar-thy-neighbor strategy of China’s even worse. The world will be awash in unsold cheap plastic parts for the next few years.

  4. Anonymous   February 10, 2009 at 6:04 pm

    It’s not 1929 for China, they are just getting started on their credit expansion. I think it’s premature to compare to Japan. When the shanghai stock index have passed 50000 points, then maybe you can worry. In many ways China are similar to Japan now, as Japan was in around 1975.