A Definitive Guide to China’s Economic Cycle Part II – New High
Summary - Shanghai surged to new high, after staying below 3300 for more than eight months (“Outlook 2017: High-Wire Act ”, 20161213). But China’s short-term growth rate cycle has peaked, and long-term cycle is declining - as indicated by our three-year property inventory investment cycle. Already, July’s economic data show that growth momentum has begun to wane. Contrary to consensus, the breakout had little to do with the revival in the old and cyclical part of the economy, but was driven by strong earnings in large caps – a structural opportunity we have been advocating. Indeed, large caps tend to outperform in an environment of moderating growth. As growth momentum is moderating rather than collapsing, the environment conducive for gains in large caps should persist, leading major stock indices to new highs. Continue to focus on earnings and size. (“An Idiot’s Guide to China’s Nifty-Fifty Run”, 20170609). And on financials, internet and tech, consumer and large cap industry leaders.
The obsession with cyclical strength stemming from an upstream resuscitation can be precarious. The surge in commodity prices is already pressuring the profitability of the mid and down-stream companies. Mathematically, such unrelenting strength in the upstream cyclical sectors means cyclicals could in theory seize all profits in the economy eventually, creating significant misallocation of resources, ultimately making the economy an entire cyclical being – a far cry from the objective of our structural reform.
The easing effect from a weak dollar is under appreciated. But most dollar bears have been buried. In a bubble, unrestrained price surge away from fundamentals is on its own self-destructive. It will also invite regulatory intervention. As the economy continues to expand, the commodity bear market rally has not exhausted. But traders’ skill will continue to be tested. Borrowing from thousands of years of Chinese ancient philosophy, we have updated our theory and models regarding China’s growth rate cycle in this note.
这是今天的报告《中国经济周期权威指南 之二 - 新高》的英文版，感谢阅读。
---------------
“Though this be madness, yet there is method in ’t.” – Hamlet, Shakespeare
For the Chinese, time is cyclic. It is a recurrence, a tapestry of monsoons, seasons and the rise and fall of dynasties. It is one of the reasons why the Chinese is prone to economic cycle theories. Since ancient times, the Chinese classics such as the I-Ching, the Records of the Grand Historian and the Discourses on Salt and Iron all had discussed cycles in the ancient agrarian economy. As commodity prices are surging back after its plunge between March and June, pundits are once again invoking economic cycle theories to point to the beginning of a “new economic cycle” – but many discussions are rich in theory and scant in data.
Regrettably, indicators from the old, more cyclical part of the economy are being applied as evidence for the “new cycle”. The term “cycle” suggests regularity, implying that economic variables fluctuate around a long-term trend with well-defined length and amplitude. In practice, it is easy to confuse interim swings with secular trends. But bullish momentum in commodity prices can be contagious. In March, we wrote the first part of “A Definitive Guide to China’s Economic Cycle”. We demonstrated the peaking of China’s three-year property inventory investment cycle - with comprehensive data. In this note, borrowing from ancient Chinese philosophy, we will discuss various Chinese cycles in further details.
The Growth Rate Cycle
The significance of any cyclical models is to identify turning points. The timing of the cycle’s peak and trough will mean profit or loss for traders. It would be futile to trade with hindsight. For forecasting purpose, our discussion will focus on the growth rate cycle – as opposed to the growth cycle that is commonly discussed. While a growth cycle consists of contraction and expansion of activity levels, a growth rate cycle measures the rate of change in economic growth, and is more adept at forecasting inflection points. The growth rate cycle can turn down, even if the level of economic activities is still expanding. Indeed, we have been applying the concept of growth rate cycle across almost all quantitative models in our writings.
Unbeknownst to many economists, the world underwent unsynchronized recessions in the 1990’s. It happened, because countries such as Germany and Japan seemed to have undergone an uninterrupted period of post-war economic growth on the surface. For these countries, the word “recession” indeed means a sustained period of below trend growth rate. Japanese growth was actually around 3% during the lost decade. China has also experienced uninterrupted growth since its “Reform and Open” in 1978. As such, the concept of growth rate cycle is more applicable when analyzing China.
The US Investment Cycle
Recently, a client approached us with a chart of US investment growth rate since 2000. The chart showed that the US investment growth rate troughed in 2002, 2009 and again in 2016, with a gap of seven years between each growth bottom (Focus Chart 1, top panel, red rectangle highlight). As the gap of seven years roughly coincides with the length of a typical Juglar Cycle of around six years, my client asked whether the current upswing in investment growth constitutes the beginning of a new cycle.
Focus Chart 1: US RoI leads GDP/investment growth, and explains the volatility in cyclical sectors well.
We believe a cycle means sustained and simultaneous upward or downward movements across key economic variables. The trend has to be pronounced, pervasive and persistent. Wesley Mitchell, an Economic Cycle guru, outlined the classic definition of business cycles as “fluctuation in aggregate economic activities”. A cycle must “consist of simultaneous expansion in many economic activities followed by similarly general recessions, (must be) recurrent … and not divisible into shorter cycles of similar magnitude and character”.
Once we extend the chart backward beyond 2000, the quasi-Juglar cycle starts to fade into the more frequent fluctuations in the US investment growth since 1947 (i.e. not recurrent). Our data show that the inception of a new investment cycle tends to begin with a substantially more negative growth rate (i.e. the cycle observed since 2000 is divisible into shorter cycles of similar magnitude and character). In short, the chart does not fit the classic definition of cycles. Indeed, the previous two bottoms of investment growth are concurrent with the US recession in 2002 and 2009. A new cycle will likely initiate with investment growth rate once again falls to its recession level.
Instead, we calculate the history of US Return on Investment (RoI) by comparing the aggregated EBITDA and Capex of US firms since 1947. We show that this measure consistently leads the US private non-residential investment by two to four quarters, and the US GDP growth by two quarters (Focus Chart 1, top and mid panels). The reason is intuitive – when investment generates good return, it attracts more investment. And vice versa.
The US investment return has been declining since 2012. It explains why US firms deploy its capital to buy back stocks, instead of reinvesting in Capex. It also explains the volatility and the bear market in commodities since 2012 (Focus Chart 1, lower panel). Indeed, this persistent decline in US RoI was the basis of our call for a severe commodity bear market back in 2013 (“Black Horse and Black Swan”, 20131207). Subsequently, oil plunged from USD100+ to USD28, iron from USD 140 to USD40 and rebar steel from RMB 3,600 to RMB1,800, before their recovery in late 2015.
The Twelve-Year Cycle
Another client approached us, with a chart showing that the growth rate of China’s industrial finished goods inventory bottomed in 2016 and started to rise. The low points of the growth rate are consistent with the levels seen in 3Q2009 and 4Q2002 – previous cyclical bottoms. Visually, the bottom in 2016 appears to be a significant low (Focus Chart 2, top panel). Thus, my client posited that the momentum of inventory restocking will likely continue.
Focus Chart 2: The 3-year property inventory cycle lags M1, coincides with rebar price and leads finished goods inventory.
In the first part of “A Definitive Guide to China’s Economic Cycle”, we discussed China’s three-year property inventory investment cycle with supporting data and charts. Our quantitative model shows that the growth rate of China’s property investment, once filtered out the noises from higher-frequency observations, demonstrates a persistent three-year cycle. This cycle is similar in length as the three-year Kitchin Inventory Cycle. Two inventory cycles make a six-year Juglar Investment Cycle. And two Juglar Cycle equates one twelve-year Kuznets Building Cycle. We then applied this 3-year Kitchin Inventory Cycle to explain the cycles in China’s other key economic variables, such as money supply, stock market, rebar steel, inflation and bond yield.
We note that the three-year property inventory cycle is peaking, or has peaked in 2Q. Already, July’s economic data have begun to disappoint. The property inventory cycle leads the cycle of industrial finished goods by around six months (Focus Chart 2, top panel). The cycle in China’s capacity utilization demonstrates a similar lead time, too (Focus Chart 3). Importantly, M1 narrow money supply growth and stock prices in turn lead the property inventory cycle by more than three months (Focus Chart 2, bottom panel).While restocking activities could extend a little further, it is already in the price (Focus Chart 4). And the momentum of restocking will wane soon, if history is a guide.
Focus Chart 3: Capacity utilization leads finished goods inventory by more than 6 months.
Focus Chart 4: The restocking is largely in the price; the momentum of stocks and metal is waning.
Ancient Chinese Philosophy about Cycles
Unfortunately, doubts arise, as there appears to be little theoretical explanations for the alleged cycles, despite clear data evidence. In our March report on China’s three-year economic cycle, we offer the three-year duration of residential building construction as a hypothesis to explain China’s property investment cycle. But nagging questions remain. Paul Samuelson once dismissed the appearance of long waves as “simply historical accidents”.
Even Kondratieff, the economist who conceived the long-wave theory, sounded hesitant when he footnoted that he had “no intention of laying the foundations for an appropriate theory of long waves” in his original paper. After all, the accuracy of cycle prediction is inversely correlated with the length of the cycle. The longer the cycle, the more vague and inscrutable the lurid details in the economic data will become. We can use inflation and growth data to divide China’s economic growth stage into four quadrants, namely, reflation, recovery, overheat and recession. We can demonstrate clearly that the Chinese economy traverse through each of these growth stages in order, from recovery to “recession” since 2009. But since 2012, it seems to be stuck in between reflation and recession, following an “L-shaped” trajectory as suggested by the “authoritative figure” last May (Focus Chart 5). China’s economic cycle has turned murky.
Focus Chart 5: China’s investment clock is stuck in “L-shape”.
The Chinese ancient philosophy is full of discussions regarding cycles long before Jevons’s sun-spot theory, which initiated the serious research into economic cycles in the west. The I-Ching, for instance, is the oldest and most profound of the ancient Chinese classics. Essentially, the I-Ching is a book of divination based on cycles. In one of its most significant diagrams, the “Qian Diagram” (乾卦, the Diagram of the Sky), the I-Ching uses different states of a dragon to signify a complete circulation in the economic cycle. For instance, “the Dragon appearing in the field” (见龙在田)refers to recovery, “the Dragon on the wing in the sky” (飞龙在天) implies prosperity and “the dragon lying hidden in the deep” (潜龙勿用) suggests depression. Until today, some Chinese research and academic institutes still incorporate the theories of the I-Ching into their economic and market forecasts.
Other Chinese classics write explicitly about a twelve-year cycle. Sima Qian’s magnum opus, “Records of the Grand Historian”, widely regarded as one of the most significant Chinese history classics, combines the Jupiter Chronology with Chinese theory of “Five Elements” to describe the agricultural cycle - from harvest, crop failure, drought, and eventually to severe famine at the end of every twelve years. “Huainanzi” describes the process of how smaller three-year cycles embedded in a larger twelve-year cycles, with “crop output declining every three years, famine happening every six year, and a serious crop failure occurring every twelve years”. “Discourses on Salt and Iron”, another Chinese ancient classic, records a similar twelve-year agricultural cycle.
These ancient Chinese classics are roughly consistent with Jevons’s eleven-year sun-spot cycle based on Heinrich Schwabe’s discovery of the solar activity cycle. As solar activities affect climate changes, the similar durations of these cycles documented in both Chinese and the western cultures are likely to be more than mere happenstances. Their duration of twelve years is consistent with four shorter, three-year property investment cycles that we have demonstrated in our own research into China’s economic cycles.
Hao Hong, CFA
2017-08-27