China Debt Dynamics: Part 2, Measuring the Burden

Rapid accumulation of debt in China is nothing new: between 2000 and 2013 the stock of bank debt rose at a compound annual average of 15.7%.  And on the face of it, that rate hasn't changed much – as of February 2014, bank lending was growing 14.2% yoy.  And that accumulation of debt has, of course, also financed dramatic economic growth: in nominal terms, GDP grew at 14.7% on a CAGR during the same period. So the question arises, what's changed? Is China's debt problem significantly worse than it used to be, and if so, what impact is it likely to have?

The purpose of this piece is simply to provide metrics which may be useful in thinking about the changing role debt finance is playing in China. It is worth, perhaps, anticipating my conclusion: the credit splurge of 2008 really did change China's financial structure, and for the worse.  The debt burden China's economy is currently carrying is dramatically more challenging than it used to be, and is certainly enough to impair current growth prospects. Moreover, as we shall see, it also means that the financial reform which China's central authorities are undertaking is absolutely necessary.  But at the same time, it poses a challenge to that reform, since the current levels of debt mean any significant rise in interest rates – which one would expect under interest-rate liberalization –  is likely be sharply punitive for those sectors encumbered with debt.

But – and this is crucial – China's cashflows as a whole remain positive, so, barring policy or political accidents,  a credit meltdown is unlikely any time soon. And that's the problem: my reading of China's shifting debt position is uncomfortable because although it clearly points to the need for change, it does not make it inevitable. China's economy can struggle on for some time under current policy-settings: that's the problem.

The fact that China's nominal GDP has grown at 14.7% CAGR this century whilst bank lending has grown at 15.7% immediately tells us that the build-up of debt/GDP has been less steep than is commonly assumed. On 2013 bank debt/GDP stood at 122%, not dramatically higher than the 116.4% of ten years earlier. China's problem lies not so much in the stock of debt per se, but rather in the falling economic efficiency of debt. One way of measuring this is to look at the amount of extra nominal GDP associated with a extra debt. One can contrast the pre and post 2009 credit-splurge periods. In 2004-2008 (the five years before the credit splurge), every 1 yuan in extra bank debt coincided with an extra 1.28 yuan in nominal GDP: in 2009-2013 (the five years after the credit splurge), evey 1 yuan in extra bank debt yields only 0.68 yuan in extra nominal GDP.  The marginal efficiency of bank lending as it related to GDP growth effectively halved.

This is the beginning of an argument not its end – one can and should question why it happened, and even whether it was a necessary adjustment. This piece is not that argument, it merely seeks to demonstrate that the economic role and impact of credit in the last five years is not the same as it was in the previous five years.

Does this declining efficiency of credit matter and if so, why? It matters because it allows us to start putting numbers on the economic burden of China's debt stock. In particular, it allows us to estimate how the financial cost of stimulating an economy by extra debt has risen. If one multiplies the average debt-stock by the average lending rate, one finds the proportion of GDP which must be committed to paying debt interest. One can then use the economic efficiency of marginal debt to calculate how much extra debt as a % of GDP would be needed to earn that amount. What this captures is the escalating effect of relying on ever greater inputs of ever-less-efficient credit in order simply to pay debt interest.

And this is where the problem really shows up: not in the rate of growth of credit per se, or in the rise in the debt/GDP ratio, but rather in the combination of these with the falling efficiency of credit. In 2004-2008 the fact that a yuan of extra credit would seemingly produce a more than an extra yuan of GDP meant that even though nominal interest rates were higher, the economic burden of servicing that debt was not increasing: taking on more credit simply made economic sense. In the 2009-2013 world, however, that has changed, with the economic burden of servicing debt more than doubling, and it making no economic sense to borrow more in order to generate the income to service that debt. 

But bank credit is no longer China's only source of finance: aggregate financing (including the 'shadow banking' which so scares commentators) has also been rising rapidly, at a CAGR of 18.7% over the last decade.  The story of declining efficiency and consequently sharply rising economic burden of the stock of financing is essentially the same as for bank credit, only in every aspect worse. I estimate the stock of aggregate financing stood at 140% of GDP in 2003 and 198% of GDP in 2013. Repeating the same calculations, the economic efficiency of aggregate financing fell from 0.8 in 2004-2008 to 0.37 in 2009-2013, and the amount of new aggregate financing needed notionally to service interest payments on the rising stock rose from 11.3% of GDP in 2004-2008 to 30.9% in 2009-2013. 

The deterioration seems extreme. So it's worth remembering three things:

1. China's private sector is still producing a sharp savings surplus, equivalent to 5.2% of GDP in 2013, I estimate. This represents (and is) the net flow of cash from the private sector into China's banks. Whilst these positive cashflows exist, systemic risks to China's banks will remain speculative.
2. Excess credit itself is likely to erode the marginal efficiency of credit, and conversely, new credit discipline can be expected to improve the efficiency of credit – and this effect can be dramatic if asset turns are forced higher.  So these numbers are less predicting an outcome than projecting a trend. And we know how well that sort of thing normally turns out. 
3. A sensible and informed China analyst will want to ask serious questions about why the efficiency of finance fell so sharply over the last five years, and will almost certainly not be fobbed off with the answer 'look at all those ghost cities.' At some stage, there are legitimate and important questions to be raised and answered about China's capital deepening. 

Nevertheless, I think the numbers do establish two things: first, the role credit plays in China's growth is fundamentally different after the 2009 credit splurge than before it; second, the scope for repeating the post-2009 credit-splurge is  limited, since  any repeat is likely to be much less effective economically and leave a much worse hangover.  China's leaders have put reform centre-stage for the next seven years, and one of the central planks of that reform is financial reform. You can see why.