In October two Chinese academics presented research proposing that China’s National Bureau of Statistics under-reports the Chinese household consumption share of GDP, officially at 36 per cent, by ten to twelve percentage points. There have been other studies suggesting that consumption is understated in the official data (although by much lower amounts). Analysts who still doubt China’s need for significant economic reform have often claimed that consumption in China is much higher than the official data reports. These upward revisions show, they argue, that the consumption imbalance in China’s economy is not as bad as is often supposed, and so will not require a sharp slowdown in economic growth as China rebalances its economy.

There are at least three reasons to be sceptical about this argument.

First, and most obviously, even the largest upward adjustments in the official data still leave China with extraordinarily low consumption. What is more, these studies ignore data that show that consumption in the official numbers might itself be overstated, possibly because most of these studies are aimed specifically at supporting the bull case.

For example, purchases of household appliances and household furnishings are among the fastest growing consumption expenditures in the official data, although it is pretty clear that they would be more usefully classified as investment.

This leads to the second reason for scepticism. What matters to China is not whether any particular expenditure is called “consumption” or “investment”. It is far more helpful to distinguish between expenditures that are highly correlated with changes in non-productive investment and those that are not.

Why? Because China’s future growth depends on the amount of domestic demand that will drive growth as Beijing brings down investment in unnecessary manufacturing capacity, infrastructure, and real estate development. If building fewer apartments reduces demand for household appliances and furnishings, then we cannot count on this “consumption” if investment in new apartment buildings declines. Or, to take another example, if purchases of gold and jewellery, among the fastest growing components of consumption, are driven by the excessively low interest rates that have also driven poor investment, the same interest rate reform that reduces investment misallocation should cause consumption of gold and jewellery to drop.

But the third and main reason for scepticism over various attempts to reclassify expenditures as consumption is that they miss the point. Countries that have had investment-led growth miracles have all, without meaningful exception, been derailed by debt. In the early stages of their growth, when investment spending is urgently needed to boost productivity, debt-servicing capacity grows faster than debt as the country pours resources into investment. This represents healthy growth.

At some stage, however, rigidities in the financial system, structural disincentives for innovation, a deficient legal system, weak education, crony capitalism, and a host of other conditions prevent further increases in capital spending from unleashing equally large increases in productivity. China has clearly passed this stage, with debt now rising much faster than debt-servicing capacity, and because Chinese growth currently is driven largely by even more rapid growth in debt, China’s problem can only be resolved by reversing this process.

If it turns out that some of the expenditures that had been previously classified as investment should in reality be classified as consumption, this does not change the debt dynamics. It only means that the same amount of debt is in fact backed by a smaller amount of investment than originally supposed. Reclassifying expenditures from investment to consumption does not make debt servicing capacity any better – after all no one would argue that debt-fuelled consumption is more sustainable than debt-fuelled investment, no matter how poor the investment.

Attempts to downplay the need for reforms by claiming that consumption numbers are understated forget that the low consumption share of China’s GDP is a consequence of a growth model in which hidden transfers from the household sector subsidize investment. With too much of this investment no longer productive, any useful analysis of China’s economy is ultimately an analysis of the relationship between credit growth and economic growth. To argue that the need for reform will be less urgent, or the adjustment will be less painful, if consumption can be shown to be higher than the official data suggest, is to misunderstand the vulnerabilities of China’s growth model.

China must find a way to grow without even faster growth in credit, and the best way to do so, as implied in the proposals following November’s Third Plenum, is to boost consumption growth by sharply increasing the household income share of GDP and to shift investment from the state sector to far more efficient smaller businesses. Beijing policy makers do not need creative ways to post higher consumption numbers; they need more efficient forms of demand. China’s problem is rapid credit growth, and the low consumption share is just a symptom of the problem.

This article was originally published in the Financial Times.