Wednesday, February 17, 2010

A Detailed Mechanism of Inflation in China, and How to Respond

EDIT: The simplest version of this story (still not super simple, but generally simpler) I wrote here: http://tfideas.blogspot.com/2010/04/nuances-around-chinas-currency.html

You may find it easier to start with this before proceeding.



The bottom half of the rest of this post is mine. The top half of the rest of this post is from a piece written by Patrick Chovanec, an economist at Tsinghua (a university in China).
http://chovanec.wordpress.com/2010/01/17/is-inflation-stalking-china/

Chovanec's is easily the best description of the potential for Chinese inflation I've ever seen. (I apologize for quoting so much, but it's really a masterpiece). I've added my own comments and extensions at the bottom.
"when the PBOC buys dollars, at a fixed exchange rate, to accumulate as reserves, it issues RMB.  Ordinarily that would mean the supply of RMB in circulation would increase, and if the economy remains the same size, each unit of RMB would buy a little bit less.  That's classic monetary price inflation, as the Monetarists themselves would describe it.

The thing is, China knows that's what will happen and wants to prevent it.  So the PBOC does two things to reduce the amount of RMB in circulation and cancel out the increase in the domestic money supply from its purchase of dollars.  First, it raises the reserve requirement of Chinese banks, and second, it issues bonds — so-called "sterilization bonds" — mainly to the banks in order to soak up even more RMB.  Josh Greenwood, Chief Economist at INVESCO, wrote an excellent paper for the Cato Institute that lays out this process in detail, for those who are interested.  But more to our immediate point, he demonstrates how together the increases in reserve requirements and total bonds issued almost exactly track China's accumulation of reserves.  The PBOC is entirely canceling out the inflationary increase in China's domestic money supply...

The funds — in the form of mandatory deposits and long-term bonds — are basically being requisitioned from China's banks, which in turn are drawing on the savings of Chinese citizens.  They do this in part by selling government bonds, in turn, to the public, and in part by investing their deposits.  In principle these are demand deposits, but in practice, for a variety of reasons — cultural predilection to save, lack of health insurance and other provisions for potential calamity — these funds are locked up in long-term savings.  And the government has a lot of leeway in tapping this large pool of savings, because (1) it owns the banks, (2) it sets interest rates, and (3) it severely restricts the range of other investment alternatives.  The effect of this borrowing is to tamp down consumer demand, not unleash it.

But as Greenwood notes, "monetary sterilization," as the process is known, is not the perfect antidote to inflation it appears to be.  Maintaining a below-market rate for the RMB, which is the whole point of the exercise, fuels the growth of the export sector...So either you get monetary inflation, by expanding the money supply, or you get "overheating" from over-incentivized exports, but one way or another you eventually get inflation."

[He cites Krugman earlier on the mechanism by which an export sector fuels inflation... I know I rail on Krugman a lot, but he is very knowledgeable about trade. The argument:
Consider the real exchange rate, defined as RX = EP*/P, where E is the exchange rate measured as the domestic currency price of foreign currency (so an appreciation of the renminbi is a fall in E), P* is the foreign price level, and P the domestic price level. Basic international macro says that there is a "natural" level of the real exchange rate, determined by trade competitiveness and international capital flows. And the economy "wants" to get to that real exchange rate.
If you have a floating exchange rate, you get there via a rise or fall in E. But if you have a pegged rate, there's pressure on prices instead. By deliberately keeping E higher than it would be under floating, China is creating pressures for P to rise; the inflationary pressures are directly related to the exchange rate policy.

However,  exports were an inflationary problem before the crisis, but since then, exports have cratered. So that's not the immediate risk. The immediate risk is the stimulus:]

"My guess — and I don't think I'm alone here — is that without the stimulus, China would pretty much be treading water on GDP.  And by "stimulus," I don't just mean the RMB 4 trillion package in government spending announced in late 2008.  I mean the more than RMB 10 trillion (USD $1.4 trillion) lending boom led by Chinese banks, of which only 2 trillion went to fund the "official" stimulus.  Even China's National Bureau of Statistics says that this lending boom dramatically expanded the money supply, in a manner that was not cancelled by any sterilization...

Where did this money go?  As I've noted before, China's lending boom evolved over the course of the year.  In Q1, it mainly went out in the form of short-term operating capital loans to prop up struggling businesses.  But evidence suggests those recipients weren't dumb.  Rather than use those funds to manufacture goods they couldn't sell, it appears many borrowers stashed the money in stocks and real estate — which helps account for the astonishing resilience of those markets in the face of last year's slowdown.

By Q2, the picture began to shift, with much of the new lending going into longer term loans, which I interpret as construction and infrastructure projects.  By Q3, the pattern continued shifting, with money going mainly into long-term consumer loans, which I take to be mortgages.  Q4 saw a moderation in lending, but with a continued emphasis on funding construction and mortgages.  There's also plenty of evidence to suggest that a big chunk of business loans to large State-Owned Enterprises (SOEs) made its way into investment in land and real estate development.

I think it's reasonable to say that — besides a big boom in automobile purchases this year — the big growth story in China this past year was construction and real estate.  The government itself, until it began to grow worried a month ago about the prospect of a property bubble, talked up these two related sectors as "key drivers" of growth.  So the picture that emerges is as follows:
  • An ongoing expansion of the RMB money supply to buy dollars (to sustain the US$ peg and support chronic trade surplus, the direct, immediate effects of which are entirely cancelled by "monetary sterilization."
  • An overstimulation export sector which, in principle, can transmit inflationary pressure to the economy, but is unlikely to have actually done so while reeling from the effects of the global slowdown for the past 14 months.
  • A massive expansion of the money supply from an unprecedented burst of bank lending that was channeled mainly into construction and real estate, and to a lesser extent the stock market, accounts for a large part of China's GDP growth over the past year, and was not counteracted by sterilization.
If it's the flow of easy money into certain sectors that's generating all or most of the GDP growth in China, it would not be surprising to find inflationary pressure — whether generated by monetary expansion or "overheating" (or perhaps both, as two faces of the same coin) — initially concentrated in those same "pockets."  You would expect to see asset inflation in real estate (as well as possibly the stock market), and price inflation in key construction inputs such as steel and cement.  Only later would those inflationary pressures potentially reach their way into the rest of the economy and affect consumer prices.  That looks strikingly like what we're seeing in the Chinese economy right now — you could call it "Crouching Stimulus, Hidden Inflation."

Of course, as overseas demand picks up, China's export sector could come back online as a source of inflationary pressure, as Krugman predicts.  But as long as China sterilizes the RMB it issues for dollars, negating the direct monetary effect of the imbalance in payments, its structural overcapacity in exports — sustained, in large part, by the stimulus — will continue to blunt that pressure for some time to come. [This refers to the fact that China has been building more factories and raw material production capacity than world demand can support. This ties into a massive bank overleverage but indicates that exports won't likely remain the problem because supply side issues are addressed]

That's not to say China should be unconcerned.  My worry is that if the expansion in construction and real estate (as well as the flow of money into the stock market), which is generating both the growth and the inflation, proves unsustainable, China could be looking at stagflation.  If you dramatically expand the money supply, and that money goes into unproductive activities that do not produce real wealth, you get the worst of both worlds — inflation without growth. "



The rest of this post is mine.

The logical extension of this discussion is "how long is Chinese growth sustainable in the nature that it has been going"?

China is targeting GDP growth through exports. It prints money to accumulate US dollar reserves in order to make its exports cheap, which increases demand for Chinese goods. The profit from exports then can go into expanding the industrial base. The industrial base is highly inefficient, cost-wise, but it is still substantial, and capacity has expanded so much that, as long as capacity can still outpace demand, I'd argue with any mechanism that suggest short- or even medium-run inflation from demand for low-price Chinese goods over high-priced foreign goods... in other words, while Krugman argues that inflation will happen with a fixed exchange rate because of price equalization, and Chovanec disputes that in the short run, saying that exports aren't high enough to drive up demand enough to move prices, I'm going to go a step farther and say that as long as capacity growth continues in the way that it has, inflation isn't going to be a problem. Additionally, there is enough excess capacity in the market that even if they don't grow capacity, inflation won't be a crushing issue for a while.

Of course, this assumes that capacity can continue to stay high. Clearly, it can't continue indefinitely. I'll come back to this.

If the US or China stops "Chinese mercantilism" and rebalances the Chinese exchange rate, demand for Chinese goods will plummet. A massive number of capital projects will default, the banks will be thrown into disarray, lending stops and WAY more stimulus is required. The Chinese economy stops growing, the money supply explodes, the country endures some lean years with unemployment skyrocketing and commodity and manufactured good deflation from a) loss of demand for capacity expansion and b) overcapacity leading to overcompetition on price, which worsens the default rate. This is a deflation trap (if this doesn't sound familiar, it should... it's not too far from what happened initially in the US, it's just that ours didn't start because of an artificially favorable exchange rate, it started because of an artificially low interest rate).

Of course, there's another step to this. To avoid a very costly deflation trap, the government prints a boatload of money and spends it - both monetary and fiscal policy (in the US, this has been Zero Interest Rates/Quantitative Easing and the dreaded S word, stimulus)... while people disagree on their importance, I think most would agree that up until now, monetary policy has been what has saved us. Either way, you're a) increasing the national debt and b) increasing the money supply.

These steps have a problem. Increased national debt almost certainly means higher interest rates, to compensate from increased default risk. An increased money supply can mean asset bubbles (anyone looking at the price of gold or treasuries?), and eventually, classic monetarist inflation, unless interest rates rise. So you end up with a situation where either a) interest rates need to rise, prolonging the downturn, or b) inflation happens.

Raising interest rates would stunt nominal growth, which would be very, very difficult to do, given their political situation, but it doesn't matter. I would speculate that in an inflation situation, the Chinese government would probably start instituting price controls domestically, and start using its trillions in US currency reserves to purchase goods internationally. The US needs to increase interest rates (nontrivially, because Chinese dollar demand will be insensitive to interest rates) or experience inflation of its own. The US starts toeing the line between stagnation, inflation or both. In a high interest rate situation, China can start using dollar reserves for international purchases and shift Yuan to transfers to people looking to finance their own consumption ("look at the Communist party/Chinese government caring for its people after the villainous West steps in to stop our growth! We will persevere!"). Either way, this doesn't look too pretty for the US.

Thus, China ends up in a pickle and real growth slows down significantly the moment the exchange rate starts slipping.

The exchange rate starts slipping when Chinese government bonds stop being appealing to Chinese citizens, and bank capital requirements can't be raised anymore because they're basically at the gold standard. Chovanec believes that stimulus will cause inflation, but stimulus funds can be mopped up with (long overdue) bank capital requirements when international demand comes back. It's possible, given the stunning size of Chinese stimulus (larger than anywhere else in the world), that China needs to suck up some inflation to absorb the stimulus (higher interest rates are barely an option at this point, because they'd slow down the growth that's propping up their asset bubbles). So yes, China could see some inflation from stimulus, and it's possible that the stimulus alone would be enough to start getting them to draw down their dollar reserves. I don't know that we have any idea how bad inflation would get, so it's hard to make that call.

What can bail out China is consumption - when the Chinese people can consume the extra capacity they have so that the government doesn't need to keep artificially pegging the exchange rate, they can stop pumping Yuan. Of course, they'd have to decrease exports at the same rate they increase consumption, or prices will start rising again, leading to interest rate hikes and recession. This brings us back to the capacity growth problem from the beginning - if capacity dips too low for export demand + consumption, prices go up, and you have interest rate hikes or inflation again. So you need consumption to grow way faster than capacity, while still leaving ample room for capacity. Functionally, what China is doing by shifting exports to consumption is "exporting their inflation" - they deflated our (US and other foreign) prices artificially, largely to our benefit, and are now walking it back down so WE pay more instead of them, and they haven't reduced consumption, so natural resources prices stay unevenly high, depending where capacity has come online.

There are a few outcomes:
1) US intervention on the exchange rate (tariffs, either of a beneficial or destructive kind) causes disaster for China, and helps the US in innumerable ways (as long as the budget deficit can come down with it), although it also causes prices to undergo a one-time spike. Spending of US currency reserves means we see inflation.

2) Unwillingness of Chinese citizens to continue financing the currency peg (but not consuming with the savings - instead, they'd be saving elsewhere), and an inability to keep forcing the banks to hold more, spells disaster for China. The currency peg slips, causing prices to undergo a one-time spike and closing the US trade deficit. Spending of US currency reserves means we see inflation.

3) China successfully walks the line by growing consumption at the same rate it walks down the exchange peg, without causing a consumption bubble. This happens before Chinese citizens aren't willing to finance the peg. Chinese growth slows down to more normal levels of productivity growth, but crisis is averted, and what a development story... meanwhile, the US undergoes a price spike, the deficit is closed, and Chinese consumption sends natural resources prices higher, leading to inflation AND the Chinese still having US monetary reserves to cause us inflation later in the future.

These can be exacerbated by:
1) Chinese stimulus causes inflation internally, leading to increased spending of US currency reserves.
2) Chinese stimulus causes only mild inflation internally, because it's sopped up by capital requirements. Their real estate bubble pops, cratering consumption.

So how does the US get out of this alive?
There are two huge reasons why the US has let China build up such huge US currency reserves.
1) We have a large government budget deficit that we have trouble financing internally because not enough people want to save government bonds. Obama has already talked about giving people tax rebates in the form of government bonds to try and finance internally. A much, much, much better solution is to cut the budget deficit. Pigouvian taxes and government spending are better targets for this for efficiency and economic growth reasons. General tax increases probably won't help the matter much.

2) We cannot provide enough oil for ourselves to run our cars, so we're ok with importing a lot of it. The money we export to buy oil ends up in Chinese hands through trade. Transitioning our auto fleet away from gasoline reduces our demand for oil.

Thus, the prescription for the US is threefold:
1) import-certificate Chinese goods. In the short run, this won't be pleasant, it may have to happen gradually, but as long as China is pegging, we don't have a choice - it just gets worse the longer we wait.
2) move vigorously to get off of gasoline in our car fleet.
3) cut government spending and implement pigouvian taxes

Simple in principle, brutal in practice, but that's how it has to go.


EDIT:
My take on things is (roughly) verified in a paper by my Introductory Macroeconomics professor, Ray Fair. http://cowles.econ.yale.edu/P/cd/d17b/d1755.pdf

An appreciation of the Yuan would make US goods more competitive with Chinese goods internally and would stimulate internal consumption. Chinese imports would drop due to recession, so there wouldn't be much help to the US there in the short run. Interest rates go up and inflation in the US happens (partially because our imports go up in price, and partially because the government can't finance its deficit). The net effect is small but we're put in a better situation of stability - as in, the sooner this happens, the less inflation happens.

Of course, what Fair doesn't consider is the idea that the government can reduce wasteful spending at the same rate as the Yuan appreciates. In that case, you'd avoid the interest rate rise, hopefully, and lessen inflation, without actually harming GDP growth much because the multiplier on wasteful government spending is way, way less than 1.


1 comment: