Thursday, October 2, 2014

The Chinese Pendulum

1. INTRODUCTION

The spectacular GDP-growth that China has managed to achieve, without a single recession over the past 30 years, has led to considerable academic and media interest in China’s economy. In particular, there has been extensive discussion and debate about the growing economic-imbalances in China. Michael Pettis, for example, has examined the sources of China’s rapid growth and analyzed its growing imbalances in great detail. Michael’s demand-side research provides clear insight into the origins of China’s growing imbalances, and suggests that China must correct these imbalances (i.e. ‘rebalance’) soon in order to avoid a debt-crisis. Michael’s research also shows that slower future GDP growth in China will be a necessary consequence of this rebalancing. 

For the convenience of the reader, here is a quick summary of Michael's outlook for China’s upcoming rebalancing-phase from his insightful demand-side analysis:
(1) Growth: There will be no crash in China. Instead, China will see a slow descent of growth-rates over a long period. GDP growth rates will fall, but they will be 'front loaded', with each passing year showing slower growth.
(2) Consumption: Consumption as share of GDP will rise, as China shifts to a consumption-led growth pattern (i.e. consumption-growth will be higher than GDP growth) in order to rebalance. As consumption-share of GDP rises, the savings-rate must necessarily fall, as they are merely the inverse of each other.
(3) Investment: Investment share of GDP will fall, as China sharply reduces investment-growth to prevent bad-debts from getting out of control and creating a banking-crisis. The investment-growth rate will be lower than the slowing GDP-growth rate and will tend to 'drag it down'.

Michael’s forecast for China is logically-sound and agrees with the general trends seen in numerous such rebalancing episodes that have occurred all over the world, ranging from Latin-America in the 1980s to Japan in the 1990s. However, we will not dwell excessively on China’s current imbalances here. Readers should visit Michael's blog for additional information on his insights into the current imbalances in China's economy.

In this article, we will instead focus on the past of China’s economic trajectory and also extend Michael’s sound demand-side research to a complementary supply-side study of events and trends.

Let us first define a non-trivial ‘imbalance’ for a developing country as a situation where consumption-share of GDP is less than 65% and falling secularly (but not necessarily monotonously) for an extended period. Alternatively, we could define it as a situation where savings-share of GDP is greater than 35% and rising secularly (but not necessarily monotonously) for an extended period. We note that savings are exactly equal to the sum of investment plus net-exports (i.e. external-surplus) as a matter of accounting identities.

Using this definition, as shown in the graph below, we can divide the Chinese economy of the past 30-years into three distinct phases:
(I) The First Imbalance-formation (1985-1994)
(II) The First Rebalancing (1994-2000)
(III) The Second Imbalance-formation (2000-Present)
The data shown in the graph above clearly indicate that the upcoming rebalancing that the whole world is expecting to see in China would actually be its second rebalancing. In order to forecast the effects of the upcoming rebalancing, therefore, we must examine the effects of its previous rebalancing in the 1990s.

2.  THE FIRST REBALANCING

A) Why did China rebalance the first time around?

The evidence indicates that investment-excesses had piled up from the first imbalance-formation phase from 1985-1994. To prevent the resulting bad-debts from overwhelming the state-owned banking system, China had no choice but to bring its run-away investment growth under control. We know that this is true, because, as shown in the graph below, China dramatically raised the bank-margins to a new high of 3.5% in order to recapitalize the banks in 1994.

B) What were the mechanisms that caused the first rebalancing?

In addition to increasing the bank-margins to recapitalize the banks, China also allowed real interest-rates to rise from 1994 onward as shown in the graph below. As Michael has shown in his research, low real interest-rates effectively act as a subsidy to investment and as a tax on consumption. This form of financial-repression is a key mechanism by which investment-led growth is created and imbalances formed in the Chinese economy. Therefore, as China raised the real interest-rate, it removed the subsidy to investment and reduced the effective tax on consumption. In effect, it reversed the interest-rates policies that had created the growing imbalance during 1985-94. Naturally, this increase in the real interest-rate caused investment growth to decline. Consumption growth moved ahead nicely, because Chinese households were now receiving much higher real returns on their bank-deposit savings. This appears to have been the principle mechanism by which the first rebalancing was achieved.

Here then is a quick summary of what happened during China’s first rebalancing during 1994-2000, and should be juxtaposed with Michael’s current-outlook for China (listed at the top) to see the obvious similarities:
 (1) Growth: There was no crash in 1994. Instead, China faced a slow descent of growth-rates over 6-year period from 1994 to 2000. Growth rates fell consistently over the period, but they were 'front loaded', with each passing year showing slower growth.
(2) Consumption: Consumption growth stayed steady even as GDP growth fell, and so consumption as share of GDP rose. In other words, China shifted to a consumption-led growth pattern order in order to rebalance. As consumption-share of GDP rose, the savings-rate necessarily fell, as they are merely the inverse of each other.
(3) Investment: Investment share of GDP fell, as China sharply reduced investment-growth to prevent bad-debts from growing out of control. The investment-growth rate was lower than the slowing GDP-growth rate and tended to 'drag it down'. In addition, China dramatically raised bank-margins to recapitalized the banks and prevent a financial crisis.

C) Why did China go back into imbalance formation after 2000?

As seen in the following graph, after having reasonably rebalanced in the nineties, China suddenly made a dramatic U-turn and rushed off at top-speed back into a state of imbalance-formation for a second time after 2000.

It is possible that the Chinese leaders felt that the bad-debt problem has been resolved and the banks adequately capitalized and so decided to return to the old ways of financial-repression generated imbalances. However, the more likely reason why China went back into massive imbalance formation after 2000 seems to have been the collapse of the internet-bubble in the United States. As seen in the following graph, unemployment took a dramatic turn upwards, probably as numerous labor-intensive export-related jobs were lost in response to the drop in consumption in the US and other countries at that time.
We note with some interest that this is the exact response we saw from China when the housing-bubble burst and the financial crisis arrived in the US in 2008. From this past evidence, it appears that Chinese policy-makers had decided long-ago that the best response to any external demand-shock is to trigger an investment-binge in order to prevent rising-unemployment and to minimize the risk of political instability.

D) How did China re-create the rising imbalances after 2000?

In order to prevent political unrest, China needed to keep unemployment from rising; and the easiest way to create lots of jobs was to trigger a construction boom. Therefore, China merely reversed the rebalancing interest-rate policy of 1994-2000 and re-introduced old financial-repression interest-rate patterns. As seen in the graph below, after peaking in 1998-99, China allowed real interest rates to drop sharply. This automatically led to rising subsidized investment-growth and a slow-down in consumption-growth, thereby taking China into its second imbalance-formation phase.

Once the imbalance-formation pattern had been re-ignited in response to the external demand-shock in 2000, the Chinese government probably ‘lost control’ (because of the ‘vested interests’) of the investment-boom and so the imbalances continued to grow even after healthy US consumption-patterns returned to the global markets after 2003-04.

Having delineated the general demand-side effects of the first rebalancing, we now turn to its supply-side effects. What effect did rebalancing itself have on unemployment? If the rebalancing had little effect on unemployment, then where was its effect felt on the supply­-side? Did it have an effect on productivity? To get the answers to these questions, we must move beyond Michael's original demand-side research and also perform a supply-side analysis of the first rebalancing in China.
.
3. SUPPLY SIDE ANALYSIS

In order to understand the mechanisms & effects of rebalancing in China, we should compare the trends seen during its first rebalancing phase (~1994-2000) to the trends seen during its second imbalance-formation phase (~2000-2006). By carefully observing the differences between the supply-side sources of growth in the two distinct phases, we can then identify the key-effects of rebalancing and then extend them to analyze China’s prospects for its upcoming second rebalancing

We will use a standard supply-side analysis as follows:

(a) Trends in total GDP*+
(b) Trends in per capita GDP*
*Note: Unlike Yen & Euro, the Chinese RMB is a pegged currency. Therefore, PPP figures have been used instead of the nominal USD ones. + Constant PPP figures are not available graphed online and so current figures have been linked as a substitute. All growth rates are calculated from PPP figures, not nominal ones.
.
The supply-side sources of this ‘growth’ are:
(i) Trends in the Productivity
(ii) Trends in the Participation-rate
(iii) Trends in the Demographic Profile
(iv) Trends in the growth of Population
http://alturl.com/mf4a7

Here are the tabulated end-point data for the first rebalancing (1994-2000) in China extracted from the World Bank data linked above: 

Similarly, here are the tabulated end-point data for the second imbalance-formation (2000-2006) extracted from the World Bank data linked above: 
Finally, here is the tabulated comparison between the average supply-side sources of growth (expressed as Cumulative Annualized Growth Rates or CAGR) over the two different phases in China:

4. DISCUSSION

4.1 Employment

The rebalancing of the 1990s clearly had negligible net-impact on unemployment. In fact, as seen in the preceding sections, it was the sudden rise in unemployment in 2000, probably caused by the collapse in external demand after the bursting of the US internet bubble, which led China to end the rebalance and go back into imbalance-formation in order to quickly create jobs. Therefore, using the example set by this precedent, we can safely say that the upcoming rebalancing in China will not lead to any significant rise in unemployment or political unrest.

In fact, given that job-seeker growth rates were higher during the 1990s (due to younger demographics) and are significantly lower now (due to demographic maturing), it follows that China will have even less of a problem with the issue of unemployment during its upcoming rebalancing that it did during its first rebalancing.

In addition, during the rebalancing of the 1990s, job-growth was preserved by means of rapid growth in the labor-intensive services sector. As seen in the figures below, the services-sector grew faster than GDP, while growth in the capital-intensive industrial-sector trailed GDP growth. As a result, the services-share of GDP grew, while the industry-share and manufacturing-share of GDP fell during the rebalancing period. This is a distinct pattern and should be expected to re-appear during China's upcoming second rebalancing.

4.2 Productivity & Labor

Since lower productivity-growth automatically leads to lower GDP-growth, it was clearly productivity that was the primary mechanism by which the slow-down required for rebalancing was achieved in the 1990s. In addition, we note that since one of the components of productivity is capital-deepening, the reduction in investment-growth during the rebalancing shows up automatically as a reduction in productivity-growth.

As seen in the preceding table, labor contributions were higher during the lower-growth rebalancing phase and actually fell during the higher-growth second imbalance-formation phase. This trend is a natural outcome of a peaking demographic profile and slowing population growth and had little to do with either rebalancing or imbalance-formation.

4.2 Capital & Efficiency

We note that ‘productivity’ has two principle components: (a) capital and (b) efficiency. Therefore, when we speak of productivity, we must distinguish as necessary between its two components. As seen in the preceding tables, the lowering of productivity-growth during the rebalancing in the 1990s was accompanied by a gain in efficiency. This implies that even though investment growth was slower during the rebalancing period, the actual investments made (i.e. capital deployed) during that period were much more judicious and productive.

Conversely, as seen in the same table, the increase in productivity-growth during the second imbalance-formation phase after 2000 was accompanied by a huge loss in efficiency. This implies that even though investment growth was faster during the imbalance-formation period, the actual investments made (i.e. capital deployed) during that period were much more indiscriminate and wasteful (i.e. malinvestment).

4.3 Quantity v/s Quality of Growth

Given that we live in a world that seems to have a growth-number fetish, it is helpful to take another look at the efficiency contributions listed in the preceding table. As seen in that table, even though output-growth was slow during the rebalancing phase in the 1990s, the associated efficiency-growth number was actually quite healthy. Conversely, even though the output-growth number increased significantly during the following imbalance-formation phase, the associated efficiency-growth number actually turned negative. A negative efficiency-growth number means that the economy is becoming less efficient (i.e. ‘wasting capital’ or ‘destroying wealth’) and generally indicates massive malinvestment.

This is the essence of the difference between the two concepts of ‘quantity of growth’ and quality of growth’. For example, slower GDP growth with rising efficiency may be better than faster GDP growth with falling efficiency. Keeping this in mind, we can conclude that even though the GDP growth number will decline as China enters into its upcoming rebalancing phase, the underlying efficiency of the economy will actually increase. This ‘rebalancing’ will transform China’s economy from a fast-growing, grossly-wasteful and unbalanced one into to a slower-growing, more-efficient and balanced one.

5. CONCLUSIONS

5.1 The Long Rebalancing

As seen in the graph below, in the first swing of the pendulum (1985-2000), we saw the imbalances grow at about 1 percentage-point of GDP a year for 9 years (1985-1994), and then we saw the rebalancing occur at about the same 1 percentage-point GDP a year for the next 6 years (1994-2000). In light of this, we could call the first rebalancing in China as the ‘Short Rebalancing’.
This time around, however, the imbalances grew a total of 12-13 percentage-points of GDP within a short-period of 6 years (2000 to 2006), giving us an annual imbalance-growth of 2 percentage-points of GDP. This implies that the second imbalance formation was twice as strong as the original one. In addition, this large imbalance of 12-13 percentage-points of GDP reached in 2006-7 has now been maintained at that level for another 6 years up to now. This implies that the second imbalance formation was also twice as long as the original one.

Given the severity of the second imbalance-formation and its extraordinary length of its persistence, we can say that the upcoming second rebalancing will be neither easy nor short. In fact, the problem is now four (= 2X2) times bigger than it was in 1994 when the first rebalancing was initiated. Therefore, the upcoming rebalancing will be long and hard, and this is why we could name it as the ‘Long Rebalancing’.

Still further, we note that once large excess investment-levels have persisted for long periods, it is only to be expected that the rebalancing will have to go beyond the usual target (35% investment) and must reduce future investment-levels even further to compensate for the excess investments made during the long and severe imbalance-formation stage in the past.

Taking into account all of the above factors, we can roughly estimate that if average GDP growth-rates (i.e. CAGR) over the entire second rebalancing phase drop to 3.5% a year (i.e. half of what was seen during the first rebalancing), then it will take about 12-14 years for China to complete its second rebalancing and arrive at a state of economy that might generously be described as ‘reasonably balanced’.

5.2 The Productivity Bubble in China

Since the analysis indicates that productivity-growth in China must slow (i.e. the productivity curve must ‘flatten’) in order for China to rebalance, we conclude that productivity has been rising in an un-natural, unsustainable and imbalanced fashion since 2000, with a wasteful reliance on excess-capital and little regard for efficiency. Therefore, what appears as malinvestment (i.e. investment-bubble) in the demand-side analysis appears as a productivity-bubble in the complementary supply-side analysis.

In light of this, here is the key curve to watch as proof of the initiation and continuance of rebalancing in China:

As seen in the graph below, the productivity curve did flatten considerably during the first rebalancing, especially between 1995 and 1999. This clear precedent then gives us the confidence to project that this ‘flattening’ of the productivity curve is indeed what will be seen again in China as it enters into its second rebalancing in the near future.
To get a feel for what a productivity curve ‘flattening’ might look like during China’s upcoming rebalancing, here is a comparison of the current productivity curves of China, Korea & Japan:
http://alturl.com/d8sba
                                 
And here is what the same curves might look like in the future (i.e. projection) if China goes into rebalancing and completes it by 2024. Observe carefully the ‘flattening’ of the productivity curve for China as its rebalancing progresses and growth-rates slow down:
In addition, here is a graph that indicates the same productivity-curve comparison expressed in ‘horizon’ form. The ‘horizon’ form shows the productivities of the trailers (or countries with lower productivity) as a percentage of the productivity of the leader (or the country with the highest productivity) and is useful in visualizing differences:

5.3 The Wild Card

Most of the patterns to be expected in the upcoming rebalancing phase in China can reasonably be projected by an extension of its own past experience with a previous rebalancing phase in the 1990s. The one wild-card in this analysis, however, is the huge housing-bubble in China. The first rebalancing phase did not have to deal with any bubble apart from the mainstream investment-bubble. Mainstream investment-bubbles do not directly affect ordinary households as they are something that corporations/SOEs, financial institutions and governments generate. The current housing-bubble, however, has directly ensnared a lot of ordinary households on the demand-side. Many households have invested large portions of their multi-generational savings into housing, either as a store of wealth like gold, or in hopes of higher-returns. Given this situation, it is not a stretch to imagine that if the housing bubble either bursts or deflates, entire such households may lose their life-time savings and so become fearful about their future. The resulting mass-fear may result in further cut backs in household consumption, thereby making the entire process of rebalancing extremely difficult. Given that was no such added-complexity during the first rebalancing, there is no precedent for us to study. Therefore, the full effects of the housing-bubble on the upcoming second rebalancing are difficult to predict and will remain a Wild Card.

5.4 The Middle Income Trap

Question: After China has finished rebalancing with slow-growth in 10-15 years, will its demographic tipping-point, which arrives around that time, push China into indefinite slow-growth like Japan? Will China necessarily get caught in a Middle-Income Trap (MIT)? Will China necessarily ‘get old before it gets rich’?

Answer: No, this is not a predestined outcome. Here is why:

We note that the stagnation in Japan in the past twenty years-- and the stagnation forecast in Korea in the next twenty years-- is a function of two things: (a) Crossing the demographic barrier (i.e. tipping over into an aging society), and (b) Reaching the productivity horizon (i.e. completing ‘catch-up’).

Even if China slows down for 10-15 years and crosses its own demographic barrier by the time it finishes rebalancing, it would still be far from the productivity horizon (i.e. it would still have some ‘catch-up’ as a source of fast-growth left). Therefore, whether China gets caught in a middle-income trap or not depends on whether China is really capable of doing ‘catch-up’ on the productivity horizon, as shown in the graph below.

We have already noted that ‘productivity’ itself is primarily a function of two things: (a) Capital availability, and (b) Efficiency (or Total-Factor Productivity TFP). Given that China, as a surplus country, has abundant capital, capital-availability should not be a debilitating factor. The issue of efficiency, however, is one that still remains to be tested in China.

Efficiency (or TFP) depends on structural, social and political reforms and refers to a variety of things such as free-competition, minimal bureaucracy, elimination or regulation of monopolies, minimizing social-friction, contract-enforcement, protection of intellectual property, promoting innovation, free-flow of information, environmental protection and so on. If China successfully implements its political, social and economic reforms, it may well be able to increase efficiency and then harness its distance from the productivity horizon as a source of fast-growth after it has completed its rebalancing. If that happens, then China need not get caught in a middle-income trap and may well progress to become a ‘rich country’ like Japan or Korea.

In summary, then, China’s fate ultimately depends on whether the Zhong Guo Gong Chan Dang is willing and able to make the reforms necessary to transform China’s system into one that is actually capable of catching-up with the ‘developed’ countries. China has the required capital and it has the required labor; but does it have the required efficiency to become a ‘developed’ country? This is the key question that remains unanswered. When it is finally answered, one way or the other, China’s pendulum will finally stop swinging. 

No comments:

Post a Comment