It has become increasingly popular in polite circles to say that China is “deleveraging”. Analysts in support of this “deleveraging” argument rely on a couple of very narrow data points that even then mangle the meaning of “deleveraging”. However, it is worthwhile to ask is China deleveraging.
Just so we all start from the same starting point, deleveraging is the process of reducing debt levels. As Wikipedia notes “It is usually measured as a decline of the total debt to GDP ratio…”. I am using Wikipedia because I want to avoid economic journals or similar technical jargon and it is a good place to start. In other words, deleveraging is generally considered a reduction in debt in either absolute or relative terms.
Allow me a brief but important tangent on what we mean by “deleveraging” in relative terms. Deleveraging in “relative terms” means the reduction in debt is not reduced in absolute terms, say I used to owe $10 now I paid back $5 and only owe $5 now. In most cases, because the denominator in nominal GDP, we are looking at whether the amount owed declines relative to national output. Put another way, does the growth rate of nominal GDP grow faster than the growth rate of debt. When you “deleverage in relative terms”, the absolute amount of debt can and normally continues to rise but just does not grow as fast as GDP.
Let us take a brief simple example. If nominal GDP and debt are both growing at 10%, there is no change in leverage relative to nominal GDP. If nominal GDP is growing at 10% and debt is growing at 15%, leverage is increasing in relative terms (15%/10%). Conversely, if nominal GDP is growing at 10% and debt is growing at 5%, leverage is decreasing in relative terms (5%/10%)<1 but continuing to increase in absolute terms. This will all come important later.
The deleveraging crowd are relying on two separate points from the Bank for International Settlements (BIS) to make their case. First, according to the BIS debt to GDP owned by non-financial corporates has slowed its growth. For instance, from Q3 2015 to Q4 2015, debt to GDP of non-financial corporates grew by 6% of GDP from 239% to 245%. However, from Q2 2016 to Q3 2017, the number slowed to 1.9% from an increase of 253.7% to 255.6%.
Second, BIS reports the “credit gap” in China has declined from a peak of 28.8 in Q1 2016 to 26.3 in Q3 2016. The BIS defines the credit gap as “as the difference between the credit-to-GDP ratio and its long-run trend” based upon the “total credit to the private non-financial sector.”
Chinese data, across a range of individual metrics, match the broad narrative that credit growth to non-financial corporates is not growing as rapidly as before. For instance, new bank loans to non-financial corporates in 2016 was down 17% to 6.1 trillion RMB. Another measure labelled Total Loans of Financial Institutions to Non-Financial Enterprises and Government Departments was up but a relatively modest 8.2% to a total of 74.1 trillion RMB. Another metric labelled Depository Corporations claims on Non-Financial Institutions was up again a modest 6.2% to 85 trillion.
Time to pop the bubble China is deleveraging right? Wrong. For obvious and non-obvious reasons. First, as you may have been able to notice by combining the data with the earlier part about how we define deleveraging, even the non-financial sector is not deleveraging in absolute or relative terms. It has only slowed the rate of adding leverage. This is like saying your breaking the speed limit by less so you should get a gold star. Neither the reduction in the credit gap nor the continued increase in debt to GDP of non-financial corporates says deleveraging. It only means that the rate of speed of additional absolute and relative leverage growth has slowed. Non-financial corporate debt to GDP isn’t leveraging up as fast but it continues to lever up.
To borrow a comparison from a previous example, if nominal GDP growth was 10% and debt growth was 15%, now nominal GDP growth is still 10% but now debt growth to non-financial corporates is only 12%. Second, what makes this search for any grain of hope to push the deleveraging story is the absolute mountain of other financial data that shows credit to other sectors exploding. If the Chinese economy was only comprised of non-financial corporates than there is hope that China would be beginning the deleveraging process. However, and this may come as a shock, there are other sectors of the Chinese economy besides non-financial corporates. While non-financial corporate debt has slowed its growth rate in excess of nominal GDP, not dropped beneath nominal GDP or gone negative, other sectors have witnessed a literal explosion of debt.
Bank loans to households were up 64% in 2016 and the first two months of 2017 they are up 75% from 2016. Nor is the household sector the insignificant after thought many make it out to be. In fact, in 2016, new bank loans to households outpaced loans to the non-financial sector 6.3 trillion RMB to 6.1 trillion RMB. Even the outstanding stock of loans to households and NFCs is closer than understood. Loans to households are a little less than half of loans to NFCs at 34 trillion to 77 trillion RMB.
The stock of household loans is up 25% since February 2016 while the stock of NFC loans is up only 8%. In short, if we take debt growth and stock from non-financial corporates in isolation, we omit one of the largest, rapid, and most important changes to Chinese credit markets. At current rates of growth, outstanding debt stock numbers will converge in about 2020 or 2021. Consequently, even if corporate credit growth slows its rate of growth, this is essentially irrelevant to the China deleveraging story.
Let me provide one more comparison that household debt is actually much larger than is realized. If we divide total household debt owed to banks by population, we arrive at a per capita debt loan of 24,903 RMB. If we use the per capita GDP number of 53,817 the household debt number does not look too bad. This gives us a household debt to per capita GDP of a solid 46%.
However, given the fact income is much lower than GDP, for a number of reasons, if we base it on the cashflows households have to pay back debt, we get a decidedly different picture. Using an urban/rural population weighting of the per capita income for urban/rural households, we produce a per capita income of 24,332. This then gives us a Chinese household per capita debt to income ratio of 102%. All of a sudden that 50% growth in new loans to households and 25% growth in the stock looks not just worrisome but downright ominous. It is worth noting this debt level does not count shadow banking products that would likely add a not insignificant amount to this number.
If we combine loans outstanding to households and non-financial enterprises and government categories, we see that outstanding loans grew 12.6% in 2016. Nominal GDP grew at 8% so the great Chinese deleveraging actually saw leverage relative to nominal GDP increase if we account for the fastest growing sector of Chinese lending. In other words, if the Chinese credit market consisted of just nonfinancial corporates and households, outstanding debt is still growing 1.59 times faster than nominal GDP.
There is one final note here. This all relies on official data and makes no assumptions about its validity. The calculations here are nothing more complicated than basic math using official numbers. However, concern about official data is perfectly valid. For instance, at the end of 2015, Liaoning would have had an official bank loan to nominal GDP ratio of 121%. However, at the end of 2016 after the National Bureau of Statistics in Beijing adjusted its GDP downward after years of self admitted fraud, this changes the outlook enormously. At the end of 2016 with the adjusted GDP data has a bank loan to nominal GDP ratio of 169%.
This is just a sliver of the overall story but even by this narrow definition, there is no deleveraging taking place even with the most generous of definitions.