The Difficult Arithmetic of Chinese Consumption

How fast does consumption need to grow in China in order for a meaningful rebalancing to take place? Probably a lot more than you think. This is arithmetically the case because China is starting from such a low base.

At roughly $1.2 trillion in 2008, total Chinese private consumption is only a little more than that of France (around $1.0 trillion) and still less than that of Germany (about $1.3 trillion, not to mention the UK’s $1.4 trillion and Japan’s $3.2 trillion). This fact alone should cause us to be extremely skeptical of feverish claims about the role Chinese consumers can play in making up for any contraction in US consumption – which at roughly $9.4 trillion last year is nearly eight times the size of China’s – without even taking into account that Europe and Japan are likely to exacerbate, rather than help absorb, the contraction in US net demand.

Chinese private consumption has dropped dramatically as a share of GDP in the past two decades. McKinsey put out a much-discussed report on consumption in August, which like many McKinsey reports is thoughtful and thorough, and generally does a good job of summarizing the informed consensus – for example the claim that a major reason for high savings is the lack of a social safety net, for which I think there is much less than meets the eye.

Unfortunately, the report tends explain the sources of low consumption too often by referring to consequences of the underlying dynamics, rather than the underlying dynamics themselves, making its proposed solutions either impractical or irrelevant. For example, the report complains that “China’s investment- and industry-intensive model crowds out consumption.”

In fact the main reason for overinvestment, and the fact that much of it is misallocated, thus widening the future gap between production and consumption, is probably too-low interest rates and a distorted credit allocation system, so it is not a question of reorienting growth away from a capital-intensive model. It requires first of all a fundamental reform of interest rate management and banking governance.

One can also easily argue that the fact that “China’s consumers make limited use of credit”, as the report claims, reflects the underlying industrial strategy more than just a technical failure to develop consumer credit. A burgeoning consumer credit market – big enough to matter – will undermine the growth model by changing the direction of implicit subsidies. This is a pretty big reform.

But that is an aside. Like most McKinsey reports it has lots of great data. For example it shows that the Chinese were not always so reluctant to consume. According to the McKinsey (and the National Bureau of Statistics) data, in 1990 consumption represented just a little over 50% of GDP. Around the time of the inflationary crisis of 1993-94 it dropped to around 45% of GDP and stayed at that level until shortly after the 1997-98 Asian crisis, when it began a fairly steep decline, hitting 40% in 2003-04 and around 35% currently. Crises seem to drive the household consumption rate down, even though bull markets don’t seem to drive it back up. Is that because crises cause households to worry about risk (although if that were true they wouldn’t go permanently down, would they)? Or is it because the government responds to crises by increasing the amount of misallocated investment, the consequence of which is to reduce future consumption? Government consumption, by the way, has stayed pretty steady, at around 15% of GDP, during that period.

Compared to non-Asian countries Chinese consumption rates are astonishingly low. Consumption for most European countries lies in the 55-65% range. Consumption for other developing countries can easily fall in the 65-70% range – where much of Latin America falls. US consumption has been around 70-72% in recent years.

Even by Asian standards Chinese consumption is off the charts. South Korean and Malaysian consumption is around 50% of GDP (although during and after the Asian crisis Malaysian consumption did drop to around 45% of GDP, before recovering). Other major Asian economies, like India, Japan, Taiwan and Thailand, show consumption in the 55-60% of GDP range. Compared to those numbers China’s 35% is astonishing, even if, as some claim it may be somewhat understated (which by the way may be true of other developing countries).

The flip side of the decline in consumption has been the rise in household savings. After bouncing around erratically between 10% and 20% of disposable income in the 1980s, around 20 years ago Chinese household savings equaled 12-15% of disposable income. Around 1992 they began rising steadily until 1998, and then stabilized at around 24-25% until very recently, when they rose slightly to about 26% of disposable income. The report correctly notes that the real increase in national savings in recent years was caused by the sharp increase in corporate savings, although as I have often mentioned before, I think corporate savings are themselves caused by the transfer from household savings via low interest rates.

During that same period China ran small surpluses or deficits on the trade account until 1996, when it booked its last trade deficit, beginning a steady upward march of its trade surplus until 2003, when the trade surplus was around 5% of GDP, after which time it surged to over 10% of GDP in 2007-2008. Investment, too, rose steadily during this period as a share of GDP. In 1990 it was around 23% of GDP. It rose sharply in 1992-94 to around 31% of GDP, stabilized at that level, and then began climbing inexorably around 1997-98 to reach around 40% in 2008.

Rising investment and rising trade surpluses are inextricably linked in China’s case. Strategies that explicitly or implicitly boosted Asian savings rates and constrained consumption, I have argued many times before, were viable strategies as long as the resulting trade surpluses, which were an almost automatic account of these policies, could be absorbed by trade deficit countries. Of course the US has played this role for the past thirty years, but there is good reason to believe that it might be able or willing to do so much longer.

These growth strategies basically forced households to subsidize investment and production, thus generating rapid economic and employment growth at the expense of household income growth, and as I have argued many times before it is the growth in household income that has primarily constrained household consumption growth.

This is borne out by the numbers. From 1990 to 2002, according to the McKinsey numbers, household income ranged from 64% of GDP to 72% of GDP. It peaked in 1992 and then began a slow, erratic descent to 66% in 2002, after which time it plunged to 55%. I suspect that if there were a way to measure changes in wealth – for example the value of the deteriorating social safety nets and the environment, the present value of savings as interest rates are changed for policy reasons, etc.—and household income were adjusted by these changes, the decline would have been greater.

The report goes on to discuss McKinsey’s projections and expectations for consumption growth over the next few years. I read it with interest but frankly I find these kinds of exercises not terribly useful because of the tremendous difficult in ascertaining the various feedback loops – of which there are many in China – which inevitably force reality far away from expectations. But I did try to do some quick arithmetic, in order to determine what kinds of numbers we are going to need to get anything resembling a rebalancing.

Rebalancing is the key word here. Many analysts think that what we need is for consumption in China to grow quickly, and this will resolve China’s (and the world’s) problem with contracting net demand in the US.

Actually, no. What we need is an expansion in Chinese net demand – rebalancing in other words – so that China can adjust to contracting net demand from the US in a way that doesn’t harm trade partners and competitors and rebounds on itself with escalating trade tensions. The way to rebalance is not for consumption to grow, but rather for consumption to grow as a share of GDP. Even if consumption declines, and GDP decline more quickly – a horrible outcome to be sure – rebalancing will occur. The best way of course is for GDP to grow quickly and for consumption to grow even more quickly.

But this is I think what most people miss. Just growth in Chinese consumption alone does not help if it grows in line with GDP, and less so if it grows slower than GDP. In that case the imbalances will get worse, and while the impact on the trade account can be temporarily disguised if investment continues to surge, ultimately it just postpones the needed adjustment (and increases the cost if the investment surge is misallocated).

What kind of consumption growth will we need for the country to rebalance? The numbers are a little worrying. If China grows by 8% a year, consumption would have to grow by a little over 11% to raise the consumption share of GDP from 35% to 36% in one year. It would have to grow by a little over 9 1/2% annually to do it in two years. Consumption, in other words, must grow substantially faster than GDP for the rebalancing even to begin to take place. This is arithmetically true because China begins the process with such a low consumption ratio.

Look at it over the longer term. Just to return consumption to 40% of GDP over the next five years (and even that level is widely considered to be way too low, and probably unprecedented in the world excluding recent Chinese history), 8% average annual growth rates in GDP would require just over 10.9% growth in consumption annually. 7% average annual GDP growth rates would require that consumption grow annually by nearly 9.9%.

In other words while GDP growth slows significantly from its 12-13% rate of the past several years, consumption will nonetheless have to surge at rates far in excess of the 8-9% growth rates of recent years in order for even a small, partial rebalancing to take place. I don’t think I have ever seen a case in which consumption has grown at nearly that rate for any length of time. I believe if China pulled it off it would be unprecedented.

Of course this will not be easy, and I think too many commentators underestimate the magnitude of the problem. China’s rebalancing process will even in the most optimistic of cases take many years before it can even reach the lowest consumption levels reached by other Asian countries that pursued investment-driven policies accompanied by too-low interest rates and undervalued currencies. This will be a long haul, and if I am right – if we need to see a transfer of income back for the state sector to the household sector really to get it going – we should expect much lower GDP growth rates over the next decade than anyone is currently projecting.

About Michael Pettis 166 Articles

Affiliation: Peking University

Michael Pettis is a professor at Peking University's Guanghua School of Management, where he specializes in Chinese financial markets. He has also 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.

Pettis has worked on Wall Street in trading, capital markets, and corporate finance since 1987, when he joined the Sovereign Debt trading team at Manufacturers Hanover (now JP Morgan). Most recently, from 1996 to 2001, Pettis worked at Bear Stearns, where he was Managing Director-Principal heading the Latin American Capital Markets and the Liability Management groups.

Visit: China Financial Markets

1 Comment on The Difficult Arithmetic of Chinese Consumption

  1. Interesting article. There are a few possible explanation on this. Krugman ( argued that China’s consumption is artificially kept low via its exchanged rate (lowered, pegged to the USD) while hold importing taxes to keep foreign products out. This results in goods being too costly within China, and so people consume less, and has a trade surplus resulting in a huge reserve of foreign capital. Another point of view was from a recent article on Money Economics ( that argued it’s part of a growth process: you need to save capital and invest in order to grow further in the future, or playing catchup in the case of China.

    There’s probably truth in both side of the arguments.

Leave a Reply

Your email address will not be published.