Why deflationary pressure in China mandates monetary tightening, not loosening

By Michael Pettis

The National Bureau of Statistics released CPI and PPI data for December 2014. People’s Daily summarizes the CPI data, which came in pretty close to market expectations:

China’s consumer prices grew 2 percent in 2014 from one year earlier, well below the government’s 3.5 percent target set for the year, official data showed on Friday. The increase was also below the 2.6-percent growth registered in 2013. Growth in the consumer price index (CPI), the main gauge of inflation, rebounded to 1.5 percent in December from November’s 1.4-percent rise, its slowest increase since November 2009. On a monthly basis, December’s CPI edged up 0.3 percent against the previous month, reversing a downward trend reported since September.

The People’s Daily also summarizes the PPI data, in which November’s 3.3% decline in prices was quite a bit worse than the 3.1% decline the market expected: PPI

China’s producer price index (PPI), which measures inflation at wholesale level, dropped 3.3 percent year on year in December, the National Bureau of Statistics said on Friday. In 2014, the country’s PPI fell 1.9 percent year on year.

Several journalists contacted me asking me to comment on the implications of the latest data, and I thought I would compile for my blog interview questions and responses from two of them. Before doing so I wanted to quote from one more People’s Daily article, in which the writer proposes very automatically a widely-held view about the monetary implications of disinflationary pressures:

China’s consumer inflation remained weak in December, while price declines at the factory gate level continued to deepen, suggesting weakness in the world’s second-largest economy but giving policy makers more room to take easing measures.

Here are the questions and my responses:

• The current data suggests that China is facing deflationary pressures, much like Japan has since the early 1990s. How will this affect the world compared to Japan’s deflation?

It will be very different. Japanese deflation occurred in an environment of fairly robust global growth. The US was just beginning the surge in productivity associated with the spread of information technology. International trade was expanding rapidly. Many developing world economies, and all of Latin America, had emerged from the terrible Lost Decade of the 1980s determined to reform and liberalize their economies. A lot of developing country debt had been written down or was in the process of being written down, and relatively speaking debt levels around the world were low and rising. Commodity prices were low, but stable, and less than a decade earlier, the Fed and other central banks around the world had been fighting off very high levels of inflation. In that environment disinflationary pressures were welcome.

Today, conditions are very different. Non-food commodity prices have declined significantly and will continue to drop a lot more. Because debt levels are extremely high everywhere many countries will be forced into deleveraging, and I suspect  it will be another five years or so before the world seriously engages in the process of restructuring sovereign debt with partial or substantial debt forgiveness. Most importantly, the two main sources of income inequality have not been resolved. First, within the household sector in the US, Europe, China, Japan and a number of other countries, the level of income inequality is nearly as bad as it has even been. Second, at the household level in countries like China and Germany the household share of income is far too low.

This combination has left us with weak consumption, excess savings and excess capacity. Without a major infrastructure investment program in the US, India, or possibly Europe, there simply isn’t enough global demand to absorb the global capacity that has been built up over the last couple of decades. So there is no appetite for disinflationary pressure in today’s global environment, whereas two decades ago the deflationary pressures that Japan might have unleashed were welcome.

I am not sure however that Chinese deflationary pressure is going to matter much to the rest of the world because China is as much a victim as it is a cause of global disinflation. The Chinese economy is simply participating in a greater global environment in which consumption is too low and the resulting excess capacity leaves the private sector unwilling to invest. But Chinese deflation is certainly not going to help.

Where China faces a problem, like many other countries, is in the relationship between debt and deflation. In a deflationary environment unless productivity growth rates are high, it is very difficult to keep the value of assets rising in line with the value of debt. There is a natural tendency for asset values to decline in line with deflation, whereas the nominal value of debt is constant (and, when interest costs are added, the nominal value of monetary obligations actually increases). Of course if the value of debt rises faster than the value of assets, by definition wealth (equal to equity, or net assets, in a corporate entity) must decline. This is why highly indebted countries and businesses struggle especially hard with deflation.

This is a problem for many Chinese borrowers. For nearly two decades, when nominal GDP growth was as high as 20-21% and the GDP deflator at 8-10%, even if they were horribly mismanaged the nominal value of assets soared relative to debt. Very low interest rate – around 7% for preferred borrowers – made servicing the debt almost an afterthought. Under those conditions it was pretty easy to ignore debt costs, and even easier to pick up very bad investment habits. Now that nominal GDP growth has dropped to around 8-10%, and could be substantially lower in a deflationary environment even if growth did not continue to decline, as I expect it will, those bad habits have become brutally expensive.

This is why borrowers are crying out for relief in the form of lower interest rates. But while lower interest rates do provide short term relief, they do not address the fundamental problem, and even allow some borrowers the lassitude to make the underlying problem worse. This puts the authorities in a tough spot.

• We are seeing reforms in many countries, especially Europe, China and Japan to open up their economies and to liberalize the labor and financial markets. When will these reforms begin to affect growth?

Unfortunately they probably won’t. Japan during the past two decades, and European countries like Spain during the past six years, should remind us very clearly of a very old story. When debt levels are low, reforms aimed at improving productivity, if they are correctly designed and implemented, can result in the higher productivity and GDP growth that could, in principle, allow a country to “grow” its way out of debt. When debt levels are high, however, reforms almost never result in faster growth. When growth is most needed, when a country is suffering from excessively high levels of debt, it is hard to find many cases in which the aggressive implementation of reforms led to growth rates fast enough for the debtor to grow its way out of debt.

This seems very counterintuitive at first, even if the history behind it is quite abundant, and very few economists seem aware of the problem (which is why most economic forecasts mistakenly focus on the pace with which reforms are likely to be implemented, and are always disappointed), but in fact the reasons are not so hard to understand. The combination of very high levels of debt and excess manufacturing capacity can lock an economy into a self-reinforcing deflationary process in which growth stagnates and debt rises faster than debt servicing capacity. When debt levels are perceived as excessive, there is downward pressure on growth for at least two reasons.

First, spending on both consumption and investment declines as households and businesses cut back on disbursements in order to repay debt (I think this is what Richard Koo refers to as “balance sheet recession”). Second, high debt levels and weak credit perceptions distort the distribution of operating earnings (at the corporate level) or the distribution of the benefits of GDP growth (at a macroeconomic level) in ways that reduce growth and increase balance sheet fragility. In finance this second reason is referred to as financial distress. By lowering growth to well below growth capacity, the combination of reduced spending and financial distress causes the real debt burden to increase. Of course an increasing debt burden reinforces the poor performance of the economy in a way familiar to anyone who has read Irving Fisher on debt-deflation.

• Comparing the causes in China and Japan, how long do you expect China will take to overcome deflation – several years, or more? What are the biggest challenges to tackle and what policies should Beijing implement?

How long it takes for China to overcome deflationary pressures depends, I think, really on two very different sets of policies. First, Beijing must aggressively tackle the country’s debt burden. For example if local governments are forced to sell off assets and use the proceeds to write down or repay debt, they can reduce the debt burden without reducing total spending. I think most policymakers and understand this, but there is another stronger reason to liquidate assets to pay down debt. Strengthening the liability side of the balance sheet changes the way assets are managed (the process is far better understood in finance theory than in economics), and the result is nearly always more productive use of the assets.

The second set of policies that Beijing should implement to protect the country from a lost decade of much slower growth is to create alternative sources of demand as quickly as possible that do not require credit expansion. I can think mainly of two ways, and both of these are implicit in the reforms proposed during the Third Plenum, in October 2013. First, substantial direct or indirect wealth transfers from the state sector to Chinese households will unleash a surge in household consumption as household income rises (and because the interest on bank deposits is an important source of income for most middle and lower middle class households, if the authorities reduce interest rates, as struggling borrowers are demanding, China actually moves in the wrong direction). The constraint here of course is political, because the elites who benefit from the state control of these assets are likely to be highly resistant to any such transfer.

Second, substantial reform in corporate governance within the banking system should be aimed at causing a substantial shift, as rapidly as possible, in the credit allocation process, so that state-related entities that systematically malinvest, like local governments and state-owned enterprises, receive a smaller share of credit while small and medium enterprises, who tend in China to be far more efficient users of capital, receive a much greater share. Of course there will also be a significant political constraint here too, and for the same reasons.

It will be hard to do either very quickly. The sets of policies that lead to either outcome are politically difficult to implement because they force a disproportionate share of the adjustment costs onto the very powerful sectors that received a disproportionately large share of China’s growth in the past two decades. What’s more, the consumption impact of wealth transfers to the household sector will lag, depending on how credible they are, while reforming the financial sector is always a slow and disruptive process.

It may take many years before China can make the necessary changes. During this time government debt will have to rise as the government absorbs the employment consequences of these disruptions, and unfortunately higher debt will itself put downward pressure on growth. It isn’t easy, but of course the history of reforms in highly indebted economies has never suggested that this would be easy, and so far it seems like Beijing is pretty determined to do whatever it has to do.

• To what extent will an acceleration in price reform help China combat disinflation?

Price reform will help much less than everyone thinks. China’s very weak consumption share of total demand has very little to do with inefficient pricing. It is almost wholly a function of the very low household share of GDP, and the only reforms that matter are reforms either that reduce the implicit transfer of wealth from households to large businesses and the state – for example allowing banks to pay higher real deposit rates, or eliminating subsidies for businesses, including land subsidies – or reforms that directly increase household wealth, including houkou reforms and improvements in the social safety net.

• As deflation also hurts Japan and Europe, what might that affect the global economy and monetary policy?

Excess capacity is a global problem, and not just a Chinese one, but the implications for monetary policy are very different in countries like China and Japan than they are in countries like Europe and the US. The monetary and financial structures of some countries create a very different set of institutions than in others, and one result is that policy responses that might seem to make sense in the US are actually harmful in China. For example lower interest rates and weaker currencies in the US and Europe might create inflationary pressure, so that the proper response to harmful deflation might very well be to reduce interest rates and to encourage currency depreciation.

It turns out, however, that under certain conditions lower interest rates and depreciating currencies may actually exacerbate deflationary pressure. Unfortunately these conditions probably apply to China and Japan. For some reason people are often shocked when I say this, even though you would have thought they would have wanted some way to explain why the roughly 35% depreciation of the yen during the last three years has not unleashed inflation, and has instead been accompanied by weaker, not stronger, consumption. Or again it should have been at the very least intriguing that during the last decade in China we have seen extraordinarily rapid monetary expansion but we have never suffered runaway CPI inflation, and in fact the inflation we have seen has been caused mainly by food shortages, not by loose money.

• How can tighter monetary policies combat deflationary pressures in Japan or China?

You get inflationary pressures when demand rises faster than supply, and deflationary pressures when the opposite happens. This is pretty easy to understand. So what matters is how monetary policies or monetary conditions affect the relationship between supply and demand. In Japan and China, especially the latter, weak consumption and high savings are not driven by very high personal savings preferences. They are driven by the low household income share of GDP. When the BoJ takes steps aimed at changing inflation expectations, for example, they are always surprised because these policies do not seem to affect Japanese psychology at all.

But in fact they probably do, it’s just that the psychology doesn’t matter. With so much being said in the press about the collapsing yen and about policies aimed at forcing up Japanese inflation, it is hard to believe that the Japanese aren’t aware of policies that are supposed to create inflation. But if there is a psychological impact, why doesn’t inflation rise?

Probably because low inflation has very little to do with Japanese household psychology. As I see it, because a weakening yen raises the cost of imports, it reduces the real value of Japanese household income while, at the same time, subsidizing the tradable goods sector. The tradable goods sector in Japan is much larger relative to the household sector than it is in the US, so perhaps it is not surprising if, unlike in the US, a weaker yen increases the growth in household income by the same amount or by less than it increases the growth in the tradable goods sector (adjusted for any change in “psychology”, of course). In that case there shouldn’t be any inflationary pressure. If consumption does not rise faster than production, after all, why should prices rise? In the end it might well take a stronger yen to force up demand relative to supply, although I suspect credibility is so low that it would take many months before the impact were felt.

Something similar happens in China, where the household income share of GDP is a much greater constraint on consumption that household psychology. In the US and Europe, deflationary pressures increase the ability of central banks to loosen monetary conditions, and because too many economists assume too easily that what is likely to be true in the US must be true everywhere, deflationary pressures in China are unleashing calls for lower interest rates and greater credit expansion in China. This is why I copied the People’s Daily article at the beginning of this blog entry.

In the US lower interest rates tend to be inflationary because a substantial portion of credit is consumer credit. What is more, lower interest rates have a positive wealth effect for American households because they tend to be associated with higher real estate prices, a stronger stock market, and of course stronger bond markets. When interest rates are lowered, the positive impact on American consumption is greater than the positive impact on American production, so prices usually rise.

In China, however, most credit is delivered to businesses, not households, and is aimed at increasing production, not consumption. What is more, for Chinese households, bank deposits form a far greater share of total financial savings than they do for American households. Lower interest rates, in other words, generally have a negative wealth effect in China largely because reducing the interest rate on bank deposits makes most Chinese feel poorer, not wealthier. An IMF study in 2011 confirmed the relationship.

This is why deflationary pressures in China indicate that we probably need monetary tightening, not loosening. I know this sounds extremely counterintuitive, and so violates what we have learned about the world by assuming that the world looks a lot like the US, but there is both a logical argument behind it and what I think is overwhelming historical evidence. The convention that any economic variable that works one way in the US must work the same way in China is one of those assumptions that is implicit in so much that is written about the Chinese economy, and yet is made by foreign and Chinese economists who would indignantly reject the assumption were it ever made explicitly.

Post script: will easing price controls cause prices to rise?

The day after I posted this entry I saw in the South China Morning Post that the weak CPI and PPI data were likely to strengthen calls for deregulating price controls. According to the article:

Fresh confirmation of persistent deflationary pressures on the mainland, with a consumer price gauge stuck near a five-year low, has prompted calls for further action to ease government price controls on energy and other key industrial inputs. Perhaps even more telling than stubbornly low consumer prices, a slide in producer prices has extended to 34 months. The entrenched factory-gate deflation might warrant a policy response of similar resolve to the leadership’s battle against corruption.

…In a sign that the delayed price reform may accelerate this year, the commission released a document on Monday for the liberalisation of tobacco leaf prices, fees charged at ports and some prices related to rail transport and civil aviation.

Price reforms in China mean, for the most part, removing price subsidies or controls that keep prices down. It may therefore seem intuitively obvious that eliminating these policies will cause average prices to rise, but this isn’t the case. China’s very weak consumption share of total demand has very little to do with inefficient pricing. It is almost wholly a function of the very low household share of GDP. Raising the prices of individual goods will not cause overall demand to rise relative to supply, and so will have no net impact on inflation.

Let’s assume, for example, that Beijing were to increase tobacco taxes so that Chinese smokers pay more for cigarettes.

1. If smokers decide in response to higher cigarette prices to reduce their savings by exactly the same amount as the higher spending on cigarettes, the result would be a one-off increase in cigarette prices and nothing else would change. This would cause a temporary jump in CPI inflation. 

2. If they spend all of their income on consumption (which is likely to be the case for poor people), or if they saved a fixed amount of their paycheck (which is likely to be the case for migrant workers and for many other types of savers), the increase in cigarette spending would be matched, kuai for kuai, with a reduction in other consumer spending. Cigarette prices would rise, but the price of other consumer goods would decline by the same amount, so that the net impact on CPI inflation would be zero.

There are perfectly good reasons to reform prices. Price reform can lead to a more efficient use of resources. If subsidies to businesses are paid indirectly or directly by households, price reforms also help rebalance the economy, which indirectly creates inflationary pressures. But it is a mistake to think that price reforms are directly inflationary. The only way to cause prices to rise is to increase demand relative to supply, and in China, where low household income is the main constraint on consumer demand, the only substantial way in which to spur CPI inflation is to increase the household income share of GDP.

 

Academics, journalists, and government and NGO officials who want to subscribe to my newsletter, which sometimes includes potions of this blog and sometimes (as in this case) does not, 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.