Thoughts about Chinese GDP, M2, and Why it Matters
- For all the hand wringing about GDP, which virtually everyone at this point recognizes has enormous flaws, it is important to understand why it matters and how to place it in the proper context. There are a couple of reasons that it matters but today we are only going to deal with one and that is the fact that GDP is used as the basis for many comparisons that we use to understand the economy. For instance, recently SoberLook pointed out the rapid increase in Chinese M2 while Simon Rabinovitch of The Economist noted that the SoberLook view point omits economic growth. Simon argues that if we correct for economic growth, then Chinese money supply nearly mirrors US M2 growth. Both are excellent points and need to be understood for what they are.
- It is very common to use GDP as a basis for a variety of indicators especially cross country comparisons and this is why GDP accuracy is so important. For instance, another common indicator people widely cite is the debt to GDP ratio. However, what if the GDP is inaccurate for either poor quality statistical work or more nefarious reasons? Due to how indicators like debt to GDP is calculated, this has an enormous impact on our understanding of the Chinese economy. Let’s use simplified but vaguely accurate numbers to illustrate the point. Assume that Chinese debt total 20 trillion RMB and the nominal value of GDP is 10 trillion RMB. We would say that China has a debt to GDP ratio of 200%. However, let’s assume that China has been fudging GDP figures every year just a little bit for the past decade, not an unrealistic assumption at all, and that nominal GDP is really only 9 trillion RMB. Now, debt to GDP has jumped from 200% to 222%. In other words, absolutely nothing has changed other than the value of the denominator but the debt to GDP ratio jumped more than 20%.
- If we return to the original point, I think there is strong evidence that GDP has systematically undercounted inflation in about the past decade. I should say, I think current prices indexes like CPI and PPI are relatively close to accurate. For instance, from my own research, the CPI urban cost of housing increased only 6% from 2000 to 2011. Let me emphasize that is not 6% annually but total. Even if we extend this forward, we would only be in the low teens for housing CPI increase between 2000 and 2015. This and other evidence implies that Chinese statistics bureau were smoothing (read hiding) inflation over the past decade something that Emi Nakamura and Jon Steinsson have also shown. This gets us back to the SoberLook and Rabinovitch comments on money and GDP growth. If we take the GDP as perfectly accurate, then we would have to believe that Chinese money growth is not necessarily excessive. However, if we make even small adjustments to GDP, again not unreasonable based upon all available evidence, many of these numbers jump enormously.
- The evidence is everywhere that money supply and the follow up lending has grown well in excessive for GDP growth. Virtually every market you could name shows serious signs of froth with the proverbial giant ball of money just rolling between asset classes. The rapid growth of the entire financial services sector that places it (as a % of GDP….notice the base) only comparable to countries like Luxembourg, Singapore, Hong Kong, or the UK which have outsized financial services sector to serve a broad international client base, something which is definitely not true of China. These are all indicators that liquidity, money, and financing has far outpaced real economic activity. A tight money environment does not produce the type of over capacity, loan growth, and asset prices we currently witness in the Chinese economy.
- Too many people when considering these questions start from the proposition that Chinese data is infallible and has to be falsified. Though I believe it is very fallible (bias alert though you likely knew that), I am contending we do not even need to falsify the data to have a better idea about what is really happening with the data and economy. Think of it this way. Assume I am a medical doctor and a patient comes to me who is obese. It is medically possible that the patient suffers from a medical condition that causes their obesity but that is far from the highest probability reason. Rather than saying what is far away the highest probability reason behind the economic outcomes we are witnessing, some still try to argue for extremely low probability medical reasons. We should look first at outcomes rather than beginning with “infallible” data.
- There are many many reasons this matters that have already mattered but let me focus on one. Money growth puts downward pressure on the currency. In fact, we are seeing this exact process playing out in China. Bond prices for junk local government debt is trading at high quality sovereign prices, real estate has rental yields of 1%, capacity throughout much of Chinese industry is at 50%, and let’s not even go down the Alice in Wonderland rabbit hole that is the stock market. What are Chinese investors going to do with their money? Try and move it out to seek better returns elsewhere and that is exactly what they are doing. Again, whether you believe the GDP and money numbers is entirely up to you, as you know my thoughts on the matter, but even if you do look at the outcomes supposed associated with nominal GDP. Is a rare medical condition causing morbid asset obesity and outflows we are witnessing? Possible yes but likely no. What is much much more likely is simply that Chinese money growth as far outpaced GDP growth which drives up asset prices and pushes money to leave.