Normally, in most countries, the distribution of income between labour and capital changes not at all or very slowly. For example, in the United States, the share of the economic pie going to workers has been, with some small exceptions, roughly stable in the post-war period. In China itself, this share was roughly stable for over 25 years since the Chinese economy took an outward turn in 1978.

But recently there have been tectonic shifts. Between 2002 and 2005, according to Berkeley economists, Chong-En Bai, Chang-Tai Hsieh, and Yingyi Qian, the share of the economic output going to workers decreased by about 8 percentage points, from about 50 per cent of GDP to 42 per cent of GDP. Which means that China -- yes, the People's Republic -- now has perhaps the lowest labour share of any major country in the world.

What does the decline in labour share mean? It does not mean that the absolute fortunes of labour have declined. To the contrary, in China, real wages have been growing at a decent clip of about 7 per cent a year. What it does imply is that real wages have been growing more slowly than productivity.

This failure of workers to capture their productivity gains -- as economic theory would predict -- has proved costly to them. If the sharing of the pie had remained the same in 2005 as in 2002, the average Chinese would have received $160 more than he or she actually did, which represents nearly 10 per cent of current per capita income.

How did this happen? Historically, such major shifts are rare. When they have occurred, they have typically been associated with political transitions as Dani Rodrik of Harvard University documented some years ago.

Transitions from democracy to autocracy (Chile in 1973, Turkey in 1980, Argentina in 1976 and Brazil in 1964) led to a large decline in the share of the pie going to labour. In fact, during these four transitions, the share of labour fell on average by 11 percentage points.

Similarly, the transition from autocracy to democracy saw an increase in the share of the pie going to labour. In a few cases -- Greece and Portugal in 1974, Spain in 1975 and Chile in 1989 -- the increase in the labour's income share was dramatic -- an average increase of about 10 percentage points. Similar changes occurred when Korea and Taiwan moved towards democracy in the 1980s.

The association between political changes and the fortunes of labour has to do with the institutional arrangements that affect the relative bargaining power of labour and capital.

Democracies tend to strengthen labour's bargaining situation, either by allowing greater freedom of association or better means of redress against employers. In some cases, democracy simply encourages greater populism, leading to large and unaffordable wage increases.

On the other hand, authoritarian regimes could favor cronyism and strengthen producer interests, resulting in a greater ability of employers to transfer income away from labour.

But China, of course, has not seen any radical political change that could easily explain the dramatic shifts that have occurred. Can market-related developments explain this puzzle? Consider the counterpart of the decline in labour share, namely the rise in the share of capital. 

It is well known, for example, that China has a distorted financial, especially banking, system, resulting in cheap and easy credit, at least to those state-owned enterprises that get it.

This distortion has if anything worsened over time. Capital has become even cheap relative to labour, contributing to rising investment, from 35 per cent of GDP in 2000 to nearly 45 per cent in 2006, and a growing capital intensity of production. With such high investment rates, it seems logical that capital's share would have risen.

Actually, standard theory suggests the opposite. The rising capital intensity of production should have reduced the returns to capital, so that the total income accruing to capital should have declined.

And this decline in turn should normally have been large enough to reduce capital's share of income or keep it unchanged. But the opposite has happened in China: the returns to capital failed to decline so that capital's share of income went up, and by a large amount. So, the puzzle deepens.

One possible explanation is technological progress. China's intensive use of capital could have been simultaneously accompanied by rapid technological progress which made capital more efficient -- or prevented it from becoming less efficient.

In fact, there's some evidence for this in the structure of exports, since over the past five years China has shifted from low-margin "commodity" manufacturing to high-margin "advanced products." But the shifts are not significant enough to warrant large changes in income distribution.

How might this decline in labour's share -- a source of potential social disaffection and unrest -- be reversed? To begin with, it is likely that public pressure will force the government to share the large returns to capital with savers, thereby improving household investment income.

Most Chinese savers, who have their money in the Chinese banking system, today obtain zero or negative returns. And they have become wise to this large disparity.

Thus, the government has been forced to list more firms in the stock market so that households can enjoy some of the high returns that companies are making. Households have also been investing heavily in the real estate market.

But this government strategy has limits because stock and real estate prices are exceptionally high, and as they return to earth, households could be left with depreciated assets and poor returns, which might do little to increase their income. 

Over a longer period, further economic forces will come into play. New entrants will emerge and bid away the excessive return to capital. But the big question is this: what if these forces are too weak, or too slow, and the public becomes impatient? Will the decline in labour's share of the economic pie be reversed through political change? That may be China's big question.

Arvind Subramanian is Senior Fellow, Peterson Institute for International Economics and Center for Global Development, and Senior Research Professor, Johns Hopkins University.