Comment 73: Is China's economy going to overheat? (December 31, 2009)

The global recession induced by the American originated financial crisis has also affected China. Its economic growth rate has declined from the overheated level of 11 plus percentage points a year to below 8 percentage points a year. Many seem to believe that China's economy is returning to the overheated bubbling stage, quoting stock market and real estate bubbles in the making. The purpose of this article is to analyze the economic conditions of China and to see whether there is the danger of overheating or not.

Before the global recession China's sizzling growth was fueled by the very strong monetary stimulus. The monetary stimulus comes from two factors, the massive influx of U. S. Dollar and the desire of the government to limit the appreciation of Yuan so that China's export boom can continue. Trade surplus, direct foreign investments, and hot money are three major sources of the incoming Dollar. All of those inbound Dollar want to be sold for Yuan. If the currency market is free, Yuan will be pushed up rapidly and China's massive export boom will be stopped. In order to sustain and expand its export stream, Chinese Government intervenes in the currency market by buying up all those unwanted Dollar to limit the appreciation of Yuan. In the process of Dollar buying Chinese Government must create matching amount of Yuan to be handed over to Dollar sellers. Thus a very large Yuan liquidity is injected into the market. The Yuan liquidity eventually flows into commercial banks. Related parties borrow those Yuan liquidity from the banks, start rampant infrastructure constructions and boost China's growth rate sky high.

As the global recession came in with full force in 2008, China's export engine was hit hard, and the direct foreign investments and the influx of hot money had stalled, too. Chinese Government bought far less Dollar and injected much less Yuan liquidity into the market. Thus the growth rate of China's economy has come down steadily. Chinese Government has started a massive stimulus program intended to uphold the economic growth rate above 8%. Then from April of 2009 something unexpected has happened. From the end of March to the end of September of 2009 China's foreign currency reserve had increased by $320 billion. As has been pointed out in Comment 72, a portion of the increase of the foreign currency reserve was due to the appreciation of Euro based assets held in the foreign currency reserve (Yen based assets are ignored here) since Euro had appreciated substantially vs. U. S. Dollar during the period. After subtracting the amount of appreciation of Euro based assets from the $320 billion increase, the remaining amount of Dollar was what Chinese Government had bought in the currency market. In Comment 72 the holding of Euro based assets was estimated to be 600 billion Euro at the end of March, a deliberate exaggeration. Even with the exaggeration, the appreciation of the Euro based assets by the end of September amounted to only $90 billion. Thus Chinese Government must have bought at least $230 billion ($320 billion - $90 billion = $230 billion) in the currency market during the 6 month period. This massive purchase of Dollar, translated to $460 billion in annual rate, far exceeded Dollar purchase at the peak of the bubble. Is the Yuan liquidity injected into the market from the massive Dollar purchase, in addition to the stimulus from the government, going to overheat Chinese economy? Before we jump into direct analysis, we need to understand more about the origins of the Dollar that had flown into China.

As mentioned before the Dollar that Chinese Government buys come from direct foreign investments, trade surplus, and hot money. Under the current global economic condition, it is doubtful that much direct foreign investment had flown into China. During the six month period from the end of March to the end of September of 2009 the aggregate trade surpluses of China was only $74 billion. Thus Chinese Government had bought $156 billion ($ 230 billion - $74 = $156 billion) from hot money. Where does the huge amount of hot money come from? How does those hot money get into China?

Hot money flows into China through gray channels but not through black markets. Let us distinguish black markets from gray channels here. Suppose someone sells Dollar to a black marketeer for Yuan. Why does the black marketeer want to buy Dollar? Apparently some one within China wants to transfer Yuan out of China. This whole transaction does not touch the official currency market that is controlled by Chinese Government. Thus the black market transactions do not become hot money and Chinese Government does not need to buy those Dollar in the black market. If a tourist to China ask a relative in China to provide Yuan and deposits the matching amount of Dollar into an oversea account of the relative, the effect is just like the black market transactions and does not become the hot money of our concern. To account for such a large amount of hot money, the most likely source is "Dollar carry trades". Dollar carry trades are to borrow Dollar at very low interest rate, convert those borrowed Dollar into Yuan in order to enjoy the future appreciation of Yuan while collecting much higher interest rate on the Yuan deposits, or to speculate on stocks or real estates. Since Chinese Government controls foreign currency transaction tightly, it takes some ingenuity to bring Dollar into China. Many businesses within China have related parties without China, like in Hong Kong. Businesses within China borrows Dollar from their outside related parties and sell those Dollar to Chinese Government for Yuan, whereas the outside related parties borrow Dollar in the open markets. Dollar carry trades exist due to the generous injection of Dollar liquidity by The Federal Reserve. Thus as long as The Federal Reserve keeps the extremely loose monetary policy, the flood of hot money into China will continue.

The effect of each Yuan injected by Chinese Government on the economy depends on the holder of the Yuan. If the holder is an exporter or a party making genuine direct foreign investment, the Yuan will be thrown back to production, and generate more earnings, spending, and savings. If the Yuan belongs to a holder of hot money or hot money disguised as direct foreign investments, it will just sit in the liquid bank accounts, in the stock markets or in the real estate markets. The latter kind of money is not an efficient way of benefiting the whole economy, though it will generate stock and real estate bubbles and enrich some speculators. To answer the question how the massive influx of hot money will or will not affect Chinese economy explicitly, we need to look deep into the hard statistical numbers published by Chinese Government.

Money markets are not well developed in China. Financial transactions in China are dominated by the central government and local government controlled commercial banks. In such an economic system money supplies become very useful in gauging the economic activity. The People's Bank of China publishes on the monthly basis three kinds of money supplies, M0, M1 and M2. M0 money supply is the amount of "currency in circulation". M1 is M0 plus all checkable deposits, and M2 is M1 plus all time deposits. The broadest money supply M2 implies the total amount of liquidity in the hands of commercial banks that can be lent out. The actual amount of liquidity that will be lent out by commercial banks depends on several factors. For example, the central bank orders commercial banks to set aside a portion of their customer's deposits as reserve, called the "required reserve", and deposit the required reserve with the central bank to reduce the lendable liquidity of the banks. The central bank may increase the required reserve to reduce the amount available for lending further in order to cool down an overheating economy. The central bank may go to opposite direction and lower the amount of required reserve to stimulate the economy. The actual amount lent out also depends on how eager borrowers are willing to borrow. The central bank can raise or lower interest rates to stimulate or suppress the eagerness of borrowing. In general when lending and borrowing are brisk, economic activities become robust, and robust economic activities will generate more earning and thus more savings. This means that M2 will grow faster when economy is growing strongly. On the other hand if the economic activity is subdued, the growth rate of M2 will slacken. Thus the observation of the growth rate of M2 is a timely method to analyze the Chinese economy.

In the graph at the right monthly values of M2, M1 and M0 in units of trillion Yuan are plotted in conventional logarithmic scales as black, red and blue curves respectively. The purple curve consists of connected line segments with turning points labeled as B, C, D, E, and F, whreas the starting point A and the ending point G are added for the convenience of the discussion. The line segment BC, for example, covers from late 2003 to late 2004, and indicates the approximate slope of the black M2 curve during the period. The slope of the line segment AB is steeper than the slope of the line segment BC. This means that the growth rate of M2 was higher during the period AB than that in the period BC so that the economic activity of China measured by M2 had peaked at the end of 2003 (point B) and had slumped until the end of 2004 (point C). The economic activity turned up from the end of 2004 (point C) and continued the robust growth until the middle of 2007 as the steeper slope of the line segment CD indicates. After the slump of the economic activity from the middle of 2007 to the middle of 2008 according to the reduced slope of the line segment DE, the activity had picked up again, but has slumped anew since the middle of 2009. What is this "economic activity" synchronized with the slope of the purple line segments, that is, the growth rate of M2? That will be our next task of the analysis.

GDP (Gross Domestic Production) tabulates the total domestic production during a set time period. There are two kinds of GDP, the nominal GDP and the real GDP. Let us consider an example to explain how those two kinds of GDP differ. Suppose a society just produces an auto mobile and a boat, and both of them do not use any foreign made parts and materials. Let us assume that the auto mobile sold to a consumer for $20,000 in 2005 and the boat sold for $80,000. The nominal GDP of the society in 2005 was $100,000, that is, $20,000 plus $80,000. Suppose in 2008 the society has produced an exactly similar auto and an exactly similar boat, but this time the boat was sold for $100,000 though the auto was sold for $20,000 without any price increase. The nominal GDP in 2008 became $100,000 plus $20,000, that is, $120,000. Thus the nominal GDP had increased by 20% from 2005 to 2008. Does this mean that the society had advanced by 20% as long as the overall production is concerned? Of course not. The society had produced same auto and boat in 2008 just as in 2005, so there was zero improvement in its ability of production. This shows that the nominal GDP is not a reliable measure to gauge the total production of the society since it is influenced by inflation. In order to correct this inflation problem the real GDP is designed. In essence the real GDP is measuring the volume of goods produced, like a car and a boat. In order to express the real GDP in one number so that its progress at different times can be compared, a proper weighting between an auto and a boat need to be introduced. To weight different products properly a reference year needs to be set. Let us choose 2005 as the reference year in our example. For 2005, the reference year, the real GDP is calculated just like the nominal GDP so it is $20,000 plus $80,000 equal to $100,000. In other words the weight of 1 to 4 for an auto vs. a boat is established. In 2008, the contribution of an auto is still $20,000 and the contribution of a boat is still $80,000, that is, their prices at the reference year, so the real GDP of 2009 is still $100,000, showing properly that the total production of the society has not improved at all. What happens if the price of an auto has gone up to $21,000 in 2008? The real GDP still remained at $100,000 since the price of an auto used in compilation is 2005 price and has nothing to do with the 2008 price. What if 2 autos are produced in 2008? Additional $20,000 needs to be added to make the real GDP $120,000. In this case the nominal GDP depends on the prices of an auto and the boat respectively. If the price of an auto is $21,000 and the price of a boat is $100,000 in 2008, the nominal GDP becomes $142,000. Since money supplies are not adjusted for inflation, the growth rate of M2 money supply should be compared with the growth rate of the nominal GDP but not with the real GDP.

China publishes its nominal GDP data in quarterly intervals but with a twist. The nominal GDP of the first quarter of each calender year is the nominal GDP number for that quarter, but the second quarter number is the sum of the first quarter and the second quarter numbers, and so on. By subtraction it is easy to reconstruct the nominal GDP for each quarter explicitly. Since Chinese Government does not perform seasonal adjustments, quarter to quarter percentage variation of the nominal GDP gyrates violently, sometimes more than 100%. Some smoothing out method must be applied to extract useful information from the data. We use year to year comparison, expressed as % change, as the way to smooth out the jumpy nominal GDP data. For example, the nominal GDP of the third quarter of 2009 had risen by 6.4% when compared to the nominal GDP of the third quarter of 2008. However, this +6.4% is not representing the condition of the third quarter of 2009 nor the condition of the third quarter of 2008. This +6.4% is rather the average condition of the middle point between the third quarter of 2009 and the third quarter of 2008, that is, the condition of the first quarter of 2009. Thus the +6.4% is plotted as a green horizontal bar under the first quarter of 2009 in the graph. We can see from the graph that every time when the growth rate of M2 slackened, the green GDP curve peaked and fell subsequently, and vice versa. From the sharp reduction of the growth rate of M2 starting from point F, we should expect that the green nominal GDP year-to-year comparison curve has formed a peak and then has declined. Of course, the nominal GDP is not the only concern. We need to look into the real GDP, too to get a comprehensive picture of Chinese economy.

Chinese Government publishes year to year percentage change of the real GDP. For example, the third quarter of 2009 number is +7.7%. This number is comparing the sum of real GDP of the first three quarters of 2009 with the sum of the first three quarters of 2008. The sum of the first three quarters of the real GDP is looking at the average condition of the second quarter of the year. Thus this +7.7% rate of increase is obtained by comparing the average condition of the second quarter of 2009 with that of the second quarter of 2008, so 7.7% growth rate was describing the average condition of the real GDP of the fourth quarter of 2008. The green nominal GDP growth rate curve at the fourth quarter of 2008 was +4.0%. The difference of 3.0% between the real and the nominal GDP means a deflation of 3.0%. The whole economic picture at the fourth quarter of 2008 is thus in spite of massive government stimulus, the increased real output of 7.7% could not be absorbed by the domestic consumption and export easily so that the overall price level was pushed down by 3.0%. If the growth rate of the nominal GDP is to come down from the third quarter of 2009 as indicated by the slower growth of M2 money supply but the real GDP should keep growing like 8.0 %, then a hefty deflation must have continued. Such an economic condition can hardly be called overheating. For example, the green curve at the second quarter of 2007 implies that the nominal GDP had grown 23% with an inflation rate about 10% to make the real GDP to grow about 13%. That is what we call "overheating".

The red M1 curve tells a different story about the Chinese economy. M1 has been synchronized with M2 most of the time until the period from F to G. The strong growth of M1 has continued whereas the growth rate of M2 has slumped since the summer of 2009. As discussed before about $150 billion hot money has flooded into China since the end of March of 2009. Translated into Yuan, it is like one trillion Yuan influx. A portion of the hot money is deposited into liquid commercial bank accounts directly and become a part of M1 immediately. Other portions of the hot money went into stock and real estate markets. However, the latter parts of the hot money just changes hands and flows into the pockets of speculators that have sold stocks or real estates, and then has been deposited into liquid bank accounts, waiting another chance to speculate. Therefore, we may consider that the hot money has directly flown into M1 money supply. From the end of March to the end of September of 2009 M1 had increased by 2.5 trillion Yuan. If 1 trillion Yuan from the hot money is subtracted, the growth rate of M1 will be similar to that of M2. This observation implies that the hot money is not an efficient agent to promote the growth of M2 and the whole economy.

The blue M0 curve in the graph, that is, the amount of currency in circulation tells another interesting story about Chinese economy. In China the usage of personal checking accounts and credit cards are not wide spread. The majority of daily transactions are made with cash. Thus the ups and downs of M0 reflect the amount of nominal consumer spending. M0 data are not adjusted for seasonal variations. The sharp peaks around each January reflect the stepped up consumer spending around the lunar new year. By looking at the slope of M0 from June to October of each year, we will get a feeling of the strength of the consumer spending of the year. For example, the consumer spending was quite weak in 2008 but has recovered in 2009 thanks to the massive governmental stimulus. Looking at the overall picture of China's economy, it seems that there are substantial problems left, but the fear of overheating is not one of them. To combat the slowing down growth rate of M2 another government stimulus may be unavoidable. In concluding this article we should note that the published GDP data is a backward looking indicator as has been pointed out already; the 7.7% growth rate of the real GDP for the third quarter of 2009 is not the growth rate of the third quarter of 2009 but is the growth rate of the fourth quarter of 2008. To know the growth rate of the real GDP of the third quarter of 2009, we need to wait until the publication of the data of the second quarter of 2010. That is why we need to study the growth rate of money supply to gain a timely indication about China's economy.