My Blog

April 4, 2008

ICBC, one of China’s Big Four commercial banks, released a report yesterday forecasting year on year CPI inflation in March to be 8.2%.  This may seem like a big improvement over February’s 8.7% inflation number, but it actually represents a significant deterioration if ICBC turns out to be right because it will nonetheless represent a big jump over the more “normal” December and January numbers (6.5% and 7.1% respectively.  February was a particularly bad month because the combined effect of the freak January storms and the Spring Festival had inflation shoot up in February by an annualized 35%.

8.2% year on year in March implies a month-on-month decline in the CPI index of about 0.8%, which after January’s and February’s 1.3% and 2.5% increases would imply that annualized inflation in the first quarter of 2008 will have just exceeded 12.5%.  I don’t have the facilities to generate my own independent projection of monthly inflation, but ICBC’s projections are in line with some of the other projections I have seen and have come to trust.  At that rate we can expect the headline year-on year CPI inflation number significantly to exceed 10% by May, which I think is a safe bet, although some analysts, for example those at ANZ, disagree – they think February’s number represents a peak for year-on-year inflation. 

I suspect that if the panic button hasn’t been pressed before then, a 10% CPI number in May should be enough to set it off, although I still think there is enough confusion in policy-making circles about the causes of inflation that it may be a while before a consensus develops over what needs to be done.  My guess is that if we do see May numbers exceed 10% there will be at best an acceleration of current policies to restrain inflation, but I don’t think these will have much effect.

In their report yesterday ICBC also said that inflation would begin to decelerate in the second half of 2008 as the government’s macro controls and support for agricultural production began to take effect.  I of course am skeptical.  I do not believe the government will be willing to tighten by nearly enough to wring out several years of excessive monetary growth.  I suspect we will need to see continuing accelerating inflation well into the end of 2008 before there is enough of a consensus built to recognize that inflation is not a temporary problem and must be addressed more vigorously, even at the cost of a short-term rise in unemployment.

Nonetheless as long as there is a sense that the fundamental problem is a food-supply constraint, the government continues to try to encourage an expansion in agricultural expansion.  Amid spreading talk around Asia of a rice crisis, Jiang Dingzhi, vice chairman of the China Banking Regulatory Commission, urged banks yesterday to ensure loans and credit to grain producers and other agriculture-related enterprises, even while they maintained strict loan caps.  In an effort that I think is unlikely to bear much fruit, Jiang even urged rural cooperatives, commercial and policy banks to “follow credit ethics and take on their social responsibilities to support rural development,” according to a Xinhua report today.  I am not sure appealing to their “better” instincts is an effective way to get the responsible parties to do what the government wants, but even if these kinds of appeals are successful in raising agricultural production, it will just make it easier for inflationary pressures to show up in the non-food component.

Why do I say this?  This month the Far East Economic Review published my piece explaining why I believe that “solving” the food problem will not solve the inflation problem.  In the article I point out that from the monthly releases provided by the NBSC since last year, it has been clear that almost all of the increase since last year in the value of the CPI can be explained by the roughly 20% increase in food prices.  The price levels of the non-food component of the CPI have been relatively stable, rising at well under 2% year on year. This seems like prima facie evidence that inflation in China is primarily, if not exclusively, a food problem, caused by food supply constraints that are temporary in nature.

So according to the pork model of Chinese inflation (based on MIT professor Ken Rogoff’s formulation of the debate as pork versus money), China is not suffering from monetary inflation. It is suffering rather from a sharp and unexpected decline in agricultural production relative to the rising food-consumption needs of a large, rapidly growing economy. Once the factors that have constrained food production are eliminated or wear off, the growth in food production will keep up with Chinese consumption needs and the price of food will return to levels consistent with the PBOC’s inflation target of 3% or lower.

For believers in the pork model the real danger facing China is that several months of high food-based inflation can nonetheless cause a generalized change in inflation expectations, which itself will change the behavior of households, consumers and producers in ways that will lock inflation into place. On February 11 Premier Wen very specifically addressed concerns about inflation expectations when he said: “We are sticking to the 4.8% target because it helps stabilize consumer expectations. When prices soar, expectations can be more horrifying than the increases.”

The concern expressed by Premier Wen and others in his government is that if workers begin demanding higher wages to compensate them for the decline in their purchasing power, if households worried by rising prices accelerate their purchases of consumer goods, and if savers reacting to the negative real interest rate on bank deposits withdraw money from the banking system and spend it, their behavior can cause inflation to spread into other sectors of the economy. In this case the main strategy of the government must be to create the necessary incentives to get agricultural production back on track as quickly as possible and to squeeze out inflationary expectations—partly by constraining demand, partly by selling food reserves, partly by freezing prices and partly by simply talking down inflation prospects, as Premier Wen was seen to be doing.

Given that CPI inflation has largely been limited to food, how can monetarists argue that inflation in China is caused by too much money, and not too little pork? Aside from the fact that the prices of a number of non-food components such as gasoline prices are frozen, so that their inflationary impact shows up not as CPI inflation but as lower profits or higher taxes (but the upward pressure nonetheless exists), there is a much more serious argument as to why non-food inflation is actually too high and possibly indicative of more generalized inflationary pressure.

In order to see why, it is important to remember that a price increase in any particular good or group of goods caused by a supply constraint is not inflationary. Prices of individual goods and services rise and fall all the time, whether the overall monetary environment is stable, inflationary or deflationary. Normally the only effect of a price rise in a particular good should be to cause a shift in relative prices, not average prices. If the price of food rises, in other words, it should cause a diversion of spending away from non-food goods and services, so putting downward pressure on the prices of those other non-food goods and services. In a perfect world, the downward pressure on other prices would net out perfectly against the rising price of food. Although there would be a change in relative prices there would be no inflation, which is a change in average prices

Of course we don’t live in a perfect world, and because of various kinds of price stickiness price changes do not necessarily net out. Still, the price-equalizing pressures are there, and it is even possible to calculate how much the price of non-food goods and services would need to decline to maintain the balance. This allows us to measure the downward pricing pressure that rising food prices are placing on the rest of the economy.

If we assume that the PBOC is running a monetary policy that is consistent with a target of 2% to 3% inflation, the 18.5% rise in food prices year on year to January 2008 would require a sharp decline in other prices in order for the PBOC to attain its inflation target. Specifically, the price of non-food goods and services would have had to decline by 5% to 6% year on year in order for overall inflation to fall within the PBOC target. (I am not using February CPI numbers because they were exceptionally high and may distort the calculation, but if we used them, non-food prices would have had to decline by 7% to 8% for the PBOC to attain its inflation target.) Food inflation, in other words, should cause such a large diversion of spending away from non-food goods and services that their prices would have to fall by nearly 5% or more if the PBOC’s monetary policy and inflation targets were credible.

It might be unreasonable to expect 5% deflation in the non-food component of the CPI basket, but certainly China does have a recent history of price deflation in many goods, and there is no reason to assume that prices in China are so sticky that deflationary adjustments are impossible.  Clothing prices in China, for example, have fallen pretty consistently year on year by an estimated 1.7%, 1.9% and 1.4% during the three months to February, according to an April 3 report just sent to me by Macquarie Bank’s Paul Cavey.  Still, even if deflation of this magnitude were unrealistic, sharply rising food prices should nonetheless have put significant downward price pressure on the non-food sector, and this downward pricing pressure should have had at least some material impact on actual price performance.

Under these circumstances the fact that non-food inflation is low but rising, and has in fact accelerated to 1.6% in February from 1.5% in January (rising nearly 3% in February on an annualized basis), is not very comforting. Instead of significant downward pressure on Chinese CPI inflation there is small, but upward pressure. This suggests that monetary policy has been too loose and that there is an underlying monetary cause to inflation. The food-supply constraint has helped mask the monetary pressure for inflation by diverting increased spending towards food and away from non-food goods and services. 

But it has only masked this pressure—it did not create it. Thanks to abnormally fast-rising food prices, in other words, inflation has not yet shown up in the non-food component because rising food prices have absorbed the inflationary consequences of an excessively loose monetary policy. Once food prices stop rising dramatically, however, inflationary pressures will show up in a much broader range of goods and services.

Comments (6) for "China's rising inflation wil...

I think in China and among bank analysts the "money" camp is much smaller than the "pork" camp, but I have to say your defense of the "money" camp is masterful and very convincing. I need to think about this a lot more and unless I can figure out where you are wrong, I may even have to switch sides. One question: what would have to happen for you to chnage your mind? By Michael Chen - 4/3/2008 6:35 PM

If the pork camp were right, prices elsewhere would be falling, because consumers paying higher food prices would have less to spend elsewhere. And if prices were falling, why would the government need to be controlling prices for energy, milk and noodles? By 8 - 4/3/2008 10:24 PM

The Chinese exporter X sells their products to the US. The USD they make are converted to RMB by the PBoC, and distributed to the workers and owners of the company X. What is the net effect. Real effect: zero (the value added has left the country). Monetary effect: RMB has been printed. Of course printing of RMB without backing of real consumable production creates inflation. Inflation is the market's "fever" to take care of the money printing. Stop the money printing, and the fever stops. By Stefan, Tallinn - 4/4/2008 7:49 PM

MC, thanks for your comments. If food prices stopped rising, or even began to fall, I would expect there to be a lot more inflation in the non-food sector as no-longer-rising food prices stopped absorbing the inflationary pressures caused by monetary expansion. This might not happen immediately, but it should show up within one or two months, so if food prices stopped rising and the inflationary pressure were not taken up by rising non-food prices within two months, I would begin seriously to question the "money" model. I suppose we will see if this is the case by late this year when, presumably, the food-supply constraints are relaxed. Of course if the "money" model is correct, by then we may well be seeing 12% year-on-year inflation or more. In that case it will be all the more difficult to stamp it out. 8, in fairness to the pork camp, sharp rises in certain food prices could easily cause the price of other foods to rise because of some kind of substitution effect. The key is not what happens to milk and noodles prices but rather prices of other non-food goods. Stefan, I agree. The trick, however, is to figure out what can be done to stop the money printing -- and whether than can be accomplished by domestic money management or by addressing the currency. By Michael Pettis - 4/4/2008 8:07 PM

Hallo 1. I count three camps: pork-camp, money-camp, global recession-camp. 2. All the big banks, that I check, have now a CPI est. 2008 >6%; they -I believe- are all going with your arguments 3. The global recession-camp (my position) has taken a hard punch: I was too naive; I did not believe, that Fed/gov. will break the rules (in that way they did) and bail out Bear Stearns; so the full collapse and cascade of exploding banks will not happen; therefore my argument for a big fall in CN-CPI to target level is now obsolete. 4. And the coming down of global GDP-growth to 3.7% (IMF) will hardly bring the CN-CPI much down: as long as the export-factories in CN do not stand still, the CPI will stay high. globumedes By globumedes - 4/4/2008 8:09 PM

As I have mentioned before when you have discussed the pork theory of Chinese inflation Michael, deflation does exist in China - in stocks. Although we have been conditioned to ignore asset prices in inflation, the theoretical case for doing so is not clear. By RebelEconomist - 4/10/2008 11:24 PM