Chinese Savings
Content
Martin Aragoneses
Paper Draft
China Macro Anomalies
Introduction
This paper argues that some of the core of the anomalies exhibited by the Chinese economy have its roots in some episodes of the recent Chinese history, having a pivotal point in 1978, the year where the economy started drifting away from central planning and embracing gradually the design of a market economy. The reforms in institutions that the generations of policymakers that started with Deng Xiao Ping changed the structure of the economy in a very particular way. Unlike other examples of transitions from communism to capitalism (e.g. the USSR), the fact that Chinese leader emphasized the need to introduce reforms in a “trial-and-error” fashion, limiting the pace of such reforms resulted in extremely high growth rates in the Chinese economy for the last thirty years. Only recently has the Chinese growth engine shown signs of fatigue, with a slowdown of GDP growth rates in the last two years, going from a two digit average growth rate over the reform period of reform down to a 7% in the 2012 estimates of the IMF? There has been wide speculation of the causes of this slowdown, and the possibility of further reduction in the pace of economic growth of the Asian giant. To be able to understand the c The third period, which spans the first decade of the new century, is a period of worldwide expansion of China and Pre Reform: Overly centralized state lack of a system of incentives to innovate. Consumption: The communist state in which Chinese economic agents lived for over thirty years made private consumption by households a mere resource to allocate robotically by the State. The focus of the Chinese government at the time was to expand
Martin Aragoneses
Paper Draft
China Macro Anomalies
investment in physical capital, sometimes, sadly, at all costs; such is the case of the illnamed “Great Leap Forward (1958-1960). The paper is structured as follows: in Chapter 1, a historical overview of the reform process serves as a vehicle for a literature review; within each sub-period of what we call Phase One of growth in China -from 1978 until the recent global financial crisis- there will be hints directed towards what will be analyzed in the next section of the paper. In Chapter 2,
Chapter 1: Historical Review of the Reform Process:
Preexisting conditions in the Maoist regime. Why do we call it Phase One. Naughton (2007) provides perhaps one of the best overviews of the post-1978 reform process in the literature. From him this paper adopts the distinction of three periods – 1978 to early 1990s, from there to the early 2000s, and basically since the accession into the World Trade Organization onwards – but with an extension, since there is a paradigm change when the 2008 financial crisis hits. Thus, the third period should end around 2008, being the forth period from then onwards. The reason why Naughton (2007) divides reform in this fashion is because of the change in interests of Chinese policy makers from one period to another. In the first period, mostly the 1980’s, the leitmotif was “growing out of the plan.” This phrase, made famous by the author, expresses the fact that during the first period of reform the goal was not to dismantle the apparatus of the centrally planned economy, but rather to let some rudimentary form of private sector grow by letting individuals
Martin Aragoneses
Paper Draft
China Macro Anomalies
experiment with the kind of incentives that rule a market economy (e.g. TVE, sell surplus at market prices, dual track price system). Thus, by a steady growth of the private sector while keeping the size of the public sector constant, Chinese policymakers succeeded in making China grow out of the plan. Chow (1993) The second period, the 1990’s is characterized by a much more aggressive play of policymakers, with a real reduction of the size of the public sector and a clear direction towards the construction of the key institutions in a market economy. The goal in mind of those in charge of the reforms was the entrance in the WTO, which required a wide range of conditions to meet by the Chinese economy with respect to liberalization of both the internal and the external market. This period was also characterized by the start of a certain degree of discontent amongst the population, which has a starting point in the Tiananmen Square protests in 1989. As Naughton (2007) puts it, this is a period of “reform with losers”. The third period goes over roughly the first decade of the new century. After China joins the WTO in Doha 2001, a new era of worldwide expansion begins. Chinese exports soar; the country’s influence in other developing areas of the world increases exponentially, giving way to what Ramo (2004) and Turin (2007)? have called the Beijing Consensus. From Angola to Reform Period: Mao Zhe Dung (commander of the second biggest centrally planned economy in the world) dies in September of 1976. System of incentives: Price change
Martin Aragoneses
Paper Draft
China Macro Anomalies
Chapter 2: Macro Anomalies of the Chinese Economy
If we were to point out the most anomalous of all of the distinctive characteristics of China’s economy, we would certainly think of the fact that the Asian giant has experiencing over the last three decades the fastest and more sustained growth period for any economy that mankind has seen. How such a rapid growth came about - certainly a fascinating topic for any economist- is one of the most studied questions in the literature. In this section, however, the analysis is focused in the underlying macroeconomic patterns that have characterized the high growth period in which the first wave of reforms have taken place (what we have called Phase One). Macroeconomic anomaly #1: the rise in savings and the fall in consumption
Consumption (%GDP) 55 Savings (%GDP)
50
45
40
35
30 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
The graph above depicts change in the composition of the Chinese GDP over the last three decades, with an increase in national savings accompanied by a decline in
Martin Aragoneses
Paper Draft
China Macro Anomalies
aggregate consumption. In 1985, consumption represented 51.3% of the GDP while savings was only 34.6%. In 2012, after the rise of savings and the fall of consumption, the numbers are reversed, being savings a 51.3% and consumption a 35.7% of the Chinese aggregate output. This major trend in the Chinese economy over the decades of reform is one of the anomalies that this paper aims to explain Argue that low sigma could be due to ageing population. Run regression of consumption on real interest rate to test it.
Thus, to estimate the elasticity one could simply regress the rate of growth of consumption on the real interest rate. (https://www.richmondfed.org/publications/research/economic_review/1989/pdf/er750601.pdf)
The real interest rate is the opportunity cost of consuming today what could be saved to obtain a return in the future. To get an intuition on why Chinese save so little, it is useful to use a standard growth model, the Ramsey-Cass-Koopmans following Ramsey (1928), Cass (1965), Koopmans (1965) model where we study how households’ savings evolve over time. The model assumes that households live for an infinite period (which is not a key assumption), and maximize their lifetime utility choosing an optimal consumption path
* ( )+given an equation of motion over time (which we assume to evolve continuously) for their savings vehicle, capital (which we denote by K). Savings in this model is what is left of income (obtained through a production function of the form () , () ( ) ( )- where AL is effective labor) after consuming and covering
the depreciation of the existing capital stock. Thus the problem of maximizing lifetime utility can be written as:
Martin Aragoneses
Paper Draft
China Macro Anomalies
* + st. ̇
∫ ( )
(
)
, ( )( )
and output per effective
( )
Here, if we define capital per effective labor labor intensive form: ̇ ( ) ( ) ( ) ( )
( ) we can then rewrite the above equation of motion in the
And we solve this problem using optimal control. We define the Present Value Hamiltonian as ( ( ) ( ) ( ))
( )
* , ( )-
( ), ( )
( )
(
) ( )-+
And using the optimality conditions for the control c(t) variable
⁄
Which becomes: ( )
, ( )( ) (that could
( )
And the optimality condition for the co-state variable intuitively thought of as the shadow price of output):
⁄
̇ ( )[ ( )
(
(
)
( ))⁄
]
( ) , ( )-( ̇ ), whose
From expression (1) we know that by the Chain Rule ̇
right hand side equated to the right hand side of (2), substituting in for (1) and dividing both sides of the expression by ( ) yields:
Martin Aragoneses
Paper Draft
China Macro Anomalies
̇ ( )
, ( ), ( ) ( ) , ( )-
-
This last expression is known as the Euler equation, and it describes the way consumption evolves optimally over time. Intuitively, consumption changes in response to differences between the discount factor for future utility, and the real interest rate
(that is the return yielded by the savings vehicle, called capital in our model). The real interest rate is simply the marginal productivity of capital (the extra output that can be produced with one extra unit of capital) net of depreciation of the existing capital stock. That is, ( ) Thus, differences between the return on savings and the time preference affect consumption, and thus, savings. Since China exhibits unusually high levels of savings and abnormally low levels of consumption, it is crucial to understand how this disparities affect income allocation, analyzing the sensitivity of savings and consumption to gaps between the real rate of return of savings and the amount of value that agents put in future utility. This is given by the elasticity of inter-temporal substitution, which we will assume to be roughly a constant denoted by . If we use a measure of instantaneous felicity of the average Chinese household that is consistent with a constant elasticity of substitution between different time periods we would have
, ( )-
, with
and
(
)
, thus
, ( )( ) , ( )And the Euler equation simplifies to
Martin Aragoneses
Paper Draft
China Macro Anomalies
̇ ( )
, ( )
will mitigate
Now it is clear that a low elasticity of intertemporal substitution,
the effect on savings of a relatively high real interest rate with respect to the discount factor for the future, and a high will make savers more responsive to this gap. Estimates with
of the elasticity of intertemporal substitution for China yield a relatively low respect to other countries. To measure if
is low of high, we compare it with the benchmark of the ( )1, for which and thus
logarithmic utility function ( )
, ( )( ) , ( )When there is an increase in the real interest rate ( ( ) ), the return on
savings increases, thus –other things equal- there is an incentive to shift consumption from today to the future (because the opportunity cost of consuming now is higher), and to increase savings. This is known as the substitution effect. But, we cannot conclude that an increase in the real interest rate leads to an increase in savings, due to the fact that there is another effect at play, which affects savings positively or negatively depending on the financial position of the average household when the exogenous increase in the interest rate occurs. In the case of the average Chinese, its position as a net saver at the time of the increase of the interest rate generates an incentive to consume more today and
1
Note that if we define
(
) ( )
( ) , applying
l’Hopital.
Martin Aragoneses
Paper Draft
China Macro Anomalies
save less since its stock of existing saving produces higher aggregate returns after the shock, which for him represents an increase in income. These two effects act as two opposite forces that drive savings up or down. The key indicator to know which one dominates is the elasticity of intertemporal substitution,
The reason why
is used as a benchmark is because it is the elasticity that
makes the income effect and substitution effect cancel out.
̇ ( )
( )
Thus in this special case, an increase in the interest rate provokes a direct increase in savings due to what is known as the wealth effect. Empirical evidence: In the case of China, all estimates indicate that elasticity of intertemporal substitution is way below one. So, in general, the income effect dominates the substitution effect when changes in the real interest rate occur.
Looking at Chinese data on the evolution of consumption, savings and real interest rates over the period where China has started to resemble a market economy, roughly from the beginning of the 90’s we find that in periods where the real interest is rising, households consume more and the country’s national savings decrease. This is one of the clear macroeconomic anomalies underlying the Chinese economy. In general, governments are able to increase consumption by decreasing the interest rate. But in China, it happens the opposite. The income effect on savers is so strong that it offsets any
Martin Aragoneses
Paper Draft
China Macro Anomalies
incentive to save more when the interest rate increases, incentivizing present consumption.
Consumption (%GDP) 55 Savings (%GDP) Interest Rate 10 8 50 6 4 Axis Title 45 2 0 40 -2 -4 35 -6 -8 30 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 -10
As can be see in the graphs, there are three periods: From 1990 to 1994, the real interest rate is decreasing while national savings increase and households’ consumption decrease. From 1994 to 1999 this trend is reversed, with soaring interest rates and falling savings, alongside with increasing consumption. Finally, from 1999 to 2008 the real interest rate falls again, increasing savings and decreasing consumption. This inverse relation between the real interest rate and savings for the Chinese economy can also be verified regressing one against the other. My rough estimates on the elasticity of intertemporal substitution of the Chinese economy is under 0.5, regressing the rate of growth of consumption on the real interest
Martin Aragoneses
Paper Draft
China Macro Anomalies
rate, as Mao (1989) suggests, following based on the work of Hansen and Singleton (1983) and Hall (1988). Following our model of savings ̇ ( )
Thus, holding time preference constant and estimating the slope parameter (how much an increase in one percentage point in the interest rate affects consumption on average) will give us an estimate for the elasticity of intertemporal substitution
6 c o n s u m t r i a o t n e g r o w t-10 h 4 2 0 -5 -2 -4 -6 -8 -10 real interest rate 0 5 10 y = 0.2675x - 2.2871
The question that emerges is why would the elasticity of inter temporal substitution be so low in China? One possible factor is the financial position of the average Chinese as a net saver. When an increase in the interest rate happens, the positive effect on the income of this average Chinese is big enough as to overflow the temptation to save more and get a higher return. Also there is trend indicating that the elasticity of intertemporal substitution is rising:
Martin Aragoneses
Paper Draft
China Macro Anomalies
EIS y1990's = 0.3155x - 1.5162
20 0 -10 -5 -20 0 5 10
EIS 2000's y = 0.4085x - 3.1081
5 0 -5 -5 -10 0 5 10
A look back to the Historical Review
Macroeconomic Anomaly #2: Excess of physical capital accumulation
Investment (%GDP)
50 48 46 44 42 40 38 36 34 32 30 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Looking at the evolution of Chinese investment as a percentage of GDP over the last thirty year shows that investment rates are always over the 35% mark, going over
Martin Aragoneses
Paper Draft
China Macro Anomalies
40% at times, and after 2008 it has surpassed the 45% threshold (being the current estimates of the IMF 48.8% of GDP). These numbers are considerably high, even for a developing economy of the size of China. It is a question that we would like to ask is it is true that the Chinese economy is over investing. From our previous section, we know that anomalies in the sensitivity of the average Chinese to changes in the interest rate accompanied by an increase in the interest rates
Macroeconomic Anomaly #3: International Competitiveness During the period of transitional dynamics that China has experienced from being a de facto closed economy in the Maoist era to becoming the giant of global production that it is nowadays, some of the most interesting features of its economy developed. Next I will specify a model for the open economy starting with the well-known identity describing GDP as the aggregate of consumption by households, investment in physical capital by firms, government expenditure and exports minus imports: ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
Where we define net exports
We have studied the evolution of savings and investment in the previous two sections. Now we turn to the differences between the two. One of the key characteristics of the Chinese economy is that it has exhibited higher savings than investment in every year since 1994, this difference is known as the Current Account balance. Intuitively it means that, since the Chinese have an excess of savings with respect to the country’s investment needs, they are able to use this surplus to lend to countries with deficits (like the United States). Similarly, the current account surplus reflects the fact that, even
Martin Aragoneses
Paper Draft
China Macro Anomalies
though China has a considerable need for imports (e.g. oil), its exporting power is far greater. This is because we can express the current account balance as: ( ) Where ( ) ( ) ( )
reflects the fact that some Chinese have financial earnings abroad (such
us in the case of holdings of stocks) that should be included in the current account surplus, and also taking care of earnings of foreigners from assets inside of China (such as profits from an American company that has a production plant in Shenzhen). This variable also accounts for the remittances that Chinese workers abroad (for example the Chinese labor force building infrastructure in Angola) send back home to their families. We see right away in the graph below that the impact of the 2007/2008 financial crisis brought down net exports due to the especially in the year 2009, when Chinese exports took a major hit, having a negative growth rate in that year for the first time since 1990. Until now (in section 1) we have relaxed the definition of savings to simply the part of output that is not consumed. But “consumption” in section 1 referred to consumption by households only. Since now our model is expanded to include the part of output that goes to government expenditure and the earnings of Chinese national abroad minus the earnings of foreigners in Mainland China, we define national savings as: ( ) ( ) ( ) ( ) ( ) ( )
and thus we can derive that the current account balance of China is equal to the excess (or deficit) of national savings over investment. First we combine identities (A) and (B):
Martin Aragoneses
Paper Draft
China Macro Anomalies
Adding (C) to both sides and subtracting Y also to both sides we get:
Exchange rate WTO accession
Conclusions: Key Macro finding #1: when the interest rate increases in China, households’ consumption goes up and national savings fall, since the substitution effect is dominated by the income effect – the Chinese have a very low elasticity of intertemporal substitution ( ) due to the facts that Chinese are net savers to begin with, and the underlying inefficiency of Chinese financial markets. Key Macro Finding #2: are the Chinese over saving? Last assembly stage, low share of valued added. Problems in technology and innovation “new growth strategies”. Naughton During the financial crisis they start taking form, policy makers start intervening into the economy, promoting the surge of credit into the economy in 2009. The crisis was crucial because it gave the policy makers the incentive to shift the development strategy. New ambitious programs are started after the crisis to boost innovation. Naughton argues that the post crisis development strategy of Chinese policy makers is a “gamble”, the government still has the power to push China into a much more higher tech development trajectory. Can it work? Yes? With better educated labor force (as it indicates the surge in college graduates) and with the huge investment made by the government, at least some of this projects will work. But Naughton argues
Martin Aragoneses
Paper Draft
China Macro Anomalies
that the large commitment of resources to this development strategy will bring inflation due to the excess liquidity in the economy. High growth high inflationary trajectory.
A WORRYING DEVELOPMENT: One of the biggest problems that higher inflation will create in China is that real interest rates will be pushed down. This is because real interest rates are defined by well-known Fisher equation: ( Where is the real interest rate, rate. Thus )( ) ( )
is expected inflation, and is the nominal interest
Since both expected inflation and the real interest rate tend to be small, we can neglect the last term in the left hand side and write a good approximation of the real interest rate as:
This last equation tells us that, effectively, if Chinese policymakers are unable to control inflation, higher inflation expectations will depress real interest rates. From our analysis of the first macroeconomic anomaly we know that, due to the low elasticity of intertemporal substitution of the average Chinese (as well as looking back at historical data), a further fall in the real interest rate is likely to make aggregate consumption decrease, aggravating the problem of lack of internal demand.
Martin Aragoneses
Paper Draft
China Macro Anomalies
Inflation 25 20 15 10 5 0 -5 -10
Real Interest Rate
The State of the Chinese Economy: Implications for China and the World (2011) video
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 12 10 8 6 4 2 0 -2 -4 -6
Current Account Balance
Martin Aragoneses
Paper Draft
China Macro Anomalies
Household Savings, the Stock Market, and Economic Growth in China
Dayong Zhang and Yu Wu 2012
Martin Aragoneses
Paper Draft
China Macro Anomalies
r & savings (90-94)
45 44 43 42 41 40 39 38 -10 -5 0 5
r & savings (94-99)
44 43 42 41 40 39 38 37 -10 -5 0 5 10
r & savings (99-08)
55 50 45 40 35 30 -5 0 5 10
Martin Aragoneses TABLE 1: Real Interest Rate (%) Period 1 1990 1991 1992 1993 1994 Period 2 1994 1995 1996 1997 1998 1999 Period 3 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 3.3 1.7 0.4 -3.6 -8 -8 -1.5 3.4 7 7.3 7.2 7.2 3.7 3.7 4.7 2.6 -1.2 1.6 2.2 -0.1 -2.3
Paper Draft
China Macro Anomalies
National Svings, Gross (%GDP) 39.216 39.366 38.774 42.542 43.573 43.573 42.118 41.287 41.827 40.188 38.191 38.191 36.831 37.581 40.302 43.827 46.832 47.964 51.518 51.845 53.351
Consumtion by Households (%GDP) 46.72893216 45.38852989 45.74304001 42.662349 41.13192118 41.13192118 42.67962511 43.5099107 43.35147337 43.95590285 45.31081121 45.31081121 46.68679083 45.65513914 43.9727904 41.84702735 40.21603567 38.09693598 35.21472793 35.95624062 34.94356141
Paper Draft
China Macro Anomalies
Introduction
This paper argues that some of the core of the anomalies exhibited by the Chinese economy have its roots in some episodes of the recent Chinese history, having a pivotal point in 1978, the year where the economy started drifting away from central planning and embracing gradually the design of a market economy. The reforms in institutions that the generations of policymakers that started with Deng Xiao Ping changed the structure of the economy in a very particular way. Unlike other examples of transitions from communism to capitalism (e.g. the USSR), the fact that Chinese leader emphasized the need to introduce reforms in a “trial-and-error” fashion, limiting the pace of such reforms resulted in extremely high growth rates in the Chinese economy for the last thirty years. Only recently has the Chinese growth engine shown signs of fatigue, with a slowdown of GDP growth rates in the last two years, going from a two digit average growth rate over the reform period of reform down to a 7% in the 2012 estimates of the IMF? There has been wide speculation of the causes of this slowdown, and the possibility of further reduction in the pace of economic growth of the Asian giant. To be able to understand the c The third period, which spans the first decade of the new century, is a period of worldwide expansion of China and Pre Reform: Overly centralized state lack of a system of incentives to innovate. Consumption: The communist state in which Chinese economic agents lived for over thirty years made private consumption by households a mere resource to allocate robotically by the State. The focus of the Chinese government at the time was to expand
Martin Aragoneses
Paper Draft
China Macro Anomalies
investment in physical capital, sometimes, sadly, at all costs; such is the case of the illnamed “Great Leap Forward (1958-1960). The paper is structured as follows: in Chapter 1, a historical overview of the reform process serves as a vehicle for a literature review; within each sub-period of what we call Phase One of growth in China -from 1978 until the recent global financial crisis- there will be hints directed towards what will be analyzed in the next section of the paper. In Chapter 2,
Chapter 1: Historical Review of the Reform Process:
Preexisting conditions in the Maoist regime. Why do we call it Phase One. Naughton (2007) provides perhaps one of the best overviews of the post-1978 reform process in the literature. From him this paper adopts the distinction of three periods – 1978 to early 1990s, from there to the early 2000s, and basically since the accession into the World Trade Organization onwards – but with an extension, since there is a paradigm change when the 2008 financial crisis hits. Thus, the third period should end around 2008, being the forth period from then onwards. The reason why Naughton (2007) divides reform in this fashion is because of the change in interests of Chinese policy makers from one period to another. In the first period, mostly the 1980’s, the leitmotif was “growing out of the plan.” This phrase, made famous by the author, expresses the fact that during the first period of reform the goal was not to dismantle the apparatus of the centrally planned economy, but rather to let some rudimentary form of private sector grow by letting individuals
Martin Aragoneses
Paper Draft
China Macro Anomalies
experiment with the kind of incentives that rule a market economy (e.g. TVE, sell surplus at market prices, dual track price system). Thus, by a steady growth of the private sector while keeping the size of the public sector constant, Chinese policymakers succeeded in making China grow out of the plan. Chow (1993) The second period, the 1990’s is characterized by a much more aggressive play of policymakers, with a real reduction of the size of the public sector and a clear direction towards the construction of the key institutions in a market economy. The goal in mind of those in charge of the reforms was the entrance in the WTO, which required a wide range of conditions to meet by the Chinese economy with respect to liberalization of both the internal and the external market. This period was also characterized by the start of a certain degree of discontent amongst the population, which has a starting point in the Tiananmen Square protests in 1989. As Naughton (2007) puts it, this is a period of “reform with losers”. The third period goes over roughly the first decade of the new century. After China joins the WTO in Doha 2001, a new era of worldwide expansion begins. Chinese exports soar; the country’s influence in other developing areas of the world increases exponentially, giving way to what Ramo (2004) and Turin (2007)? have called the Beijing Consensus. From Angola to Reform Period: Mao Zhe Dung (commander of the second biggest centrally planned economy in the world) dies in September of 1976. System of incentives: Price change
Martin Aragoneses
Paper Draft
China Macro Anomalies
Chapter 2: Macro Anomalies of the Chinese Economy
If we were to point out the most anomalous of all of the distinctive characteristics of China’s economy, we would certainly think of the fact that the Asian giant has experiencing over the last three decades the fastest and more sustained growth period for any economy that mankind has seen. How such a rapid growth came about - certainly a fascinating topic for any economist- is one of the most studied questions in the literature. In this section, however, the analysis is focused in the underlying macroeconomic patterns that have characterized the high growth period in which the first wave of reforms have taken place (what we have called Phase One). Macroeconomic anomaly #1: the rise in savings and the fall in consumption
Consumption (%GDP) 55 Savings (%GDP)
50
45
40
35
30 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
The graph above depicts change in the composition of the Chinese GDP over the last three decades, with an increase in national savings accompanied by a decline in
Martin Aragoneses
Paper Draft
China Macro Anomalies
aggregate consumption. In 1985, consumption represented 51.3% of the GDP while savings was only 34.6%. In 2012, after the rise of savings and the fall of consumption, the numbers are reversed, being savings a 51.3% and consumption a 35.7% of the Chinese aggregate output. This major trend in the Chinese economy over the decades of reform is one of the anomalies that this paper aims to explain Argue that low sigma could be due to ageing population. Run regression of consumption on real interest rate to test it.
Thus, to estimate the elasticity one could simply regress the rate of growth of consumption on the real interest rate. (https://www.richmondfed.org/publications/research/economic_review/1989/pdf/er750601.pdf)
The real interest rate is the opportunity cost of consuming today what could be saved to obtain a return in the future. To get an intuition on why Chinese save so little, it is useful to use a standard growth model, the Ramsey-Cass-Koopmans following Ramsey (1928), Cass (1965), Koopmans (1965) model where we study how households’ savings evolve over time. The model assumes that households live for an infinite period (which is not a key assumption), and maximize their lifetime utility choosing an optimal consumption path
* ( )+given an equation of motion over time (which we assume to evolve continuously) for their savings vehicle, capital (which we denote by K). Savings in this model is what is left of income (obtained through a production function of the form () , () ( ) ( )- where AL is effective labor) after consuming and covering
the depreciation of the existing capital stock. Thus the problem of maximizing lifetime utility can be written as:
Martin Aragoneses
Paper Draft
China Macro Anomalies
* + st. ̇
∫ ( )
(
)
, ( )( )
and output per effective
( )
Here, if we define capital per effective labor labor intensive form: ̇ ( ) ( ) ( ) ( )
( ) we can then rewrite the above equation of motion in the
And we solve this problem using optimal control. We define the Present Value Hamiltonian as ( ( ) ( ) ( ))
( )
* , ( )-
( ), ( )
( )
(
) ( )-+
And using the optimality conditions for the control c(t) variable
⁄
Which becomes: ( )
, ( )( ) (that could
( )
And the optimality condition for the co-state variable intuitively thought of as the shadow price of output):
⁄
̇ ( )[ ( )
(
(
)
( ))⁄
]
( ) , ( )-( ̇ ), whose
From expression (1) we know that by the Chain Rule ̇
right hand side equated to the right hand side of (2), substituting in for (1) and dividing both sides of the expression by ( ) yields:
Martin Aragoneses
Paper Draft
China Macro Anomalies
̇ ( )
, ( ), ( ) ( ) , ( )-
-
This last expression is known as the Euler equation, and it describes the way consumption evolves optimally over time. Intuitively, consumption changes in response to differences between the discount factor for future utility, and the real interest rate
(that is the return yielded by the savings vehicle, called capital in our model). The real interest rate is simply the marginal productivity of capital (the extra output that can be produced with one extra unit of capital) net of depreciation of the existing capital stock. That is, ( ) Thus, differences between the return on savings and the time preference affect consumption, and thus, savings. Since China exhibits unusually high levels of savings and abnormally low levels of consumption, it is crucial to understand how this disparities affect income allocation, analyzing the sensitivity of savings and consumption to gaps between the real rate of return of savings and the amount of value that agents put in future utility. This is given by the elasticity of inter-temporal substitution, which we will assume to be roughly a constant denoted by . If we use a measure of instantaneous felicity of the average Chinese household that is consistent with a constant elasticity of substitution between different time periods we would have
, ( )-
, with
and
(
)
, thus
, ( )( ) , ( )And the Euler equation simplifies to
Martin Aragoneses
Paper Draft
China Macro Anomalies
̇ ( )
, ( )
will mitigate
Now it is clear that a low elasticity of intertemporal substitution,
the effect on savings of a relatively high real interest rate with respect to the discount factor for the future, and a high will make savers more responsive to this gap. Estimates with
of the elasticity of intertemporal substitution for China yield a relatively low respect to other countries. To measure if
is low of high, we compare it with the benchmark of the ( )1, for which and thus
logarithmic utility function ( )
, ( )( ) , ( )When there is an increase in the real interest rate ( ( ) ), the return on
savings increases, thus –other things equal- there is an incentive to shift consumption from today to the future (because the opportunity cost of consuming now is higher), and to increase savings. This is known as the substitution effect. But, we cannot conclude that an increase in the real interest rate leads to an increase in savings, due to the fact that there is another effect at play, which affects savings positively or negatively depending on the financial position of the average household when the exogenous increase in the interest rate occurs. In the case of the average Chinese, its position as a net saver at the time of the increase of the interest rate generates an incentive to consume more today and
1
Note that if we define
(
) ( )
( ) , applying
l’Hopital.
Martin Aragoneses
Paper Draft
China Macro Anomalies
save less since its stock of existing saving produces higher aggregate returns after the shock, which for him represents an increase in income. These two effects act as two opposite forces that drive savings up or down. The key indicator to know which one dominates is the elasticity of intertemporal substitution,
The reason why
is used as a benchmark is because it is the elasticity that
makes the income effect and substitution effect cancel out.
̇ ( )
( )
Thus in this special case, an increase in the interest rate provokes a direct increase in savings due to what is known as the wealth effect. Empirical evidence: In the case of China, all estimates indicate that elasticity of intertemporal substitution is way below one. So, in general, the income effect dominates the substitution effect when changes in the real interest rate occur.
Looking at Chinese data on the evolution of consumption, savings and real interest rates over the period where China has started to resemble a market economy, roughly from the beginning of the 90’s we find that in periods where the real interest is rising, households consume more and the country’s national savings decrease. This is one of the clear macroeconomic anomalies underlying the Chinese economy. In general, governments are able to increase consumption by decreasing the interest rate. But in China, it happens the opposite. The income effect on savers is so strong that it offsets any
Martin Aragoneses
Paper Draft
China Macro Anomalies
incentive to save more when the interest rate increases, incentivizing present consumption.
Consumption (%GDP) 55 Savings (%GDP) Interest Rate 10 8 50 6 4 Axis Title 45 2 0 40 -2 -4 35 -6 -8 30 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 -10
As can be see in the graphs, there are three periods: From 1990 to 1994, the real interest rate is decreasing while national savings increase and households’ consumption decrease. From 1994 to 1999 this trend is reversed, with soaring interest rates and falling savings, alongside with increasing consumption. Finally, from 1999 to 2008 the real interest rate falls again, increasing savings and decreasing consumption. This inverse relation between the real interest rate and savings for the Chinese economy can also be verified regressing one against the other. My rough estimates on the elasticity of intertemporal substitution of the Chinese economy is under 0.5, regressing the rate of growth of consumption on the real interest
Martin Aragoneses
Paper Draft
China Macro Anomalies
rate, as Mao (1989) suggests, following based on the work of Hansen and Singleton (1983) and Hall (1988). Following our model of savings ̇ ( )
Thus, holding time preference constant and estimating the slope parameter (how much an increase in one percentage point in the interest rate affects consumption on average) will give us an estimate for the elasticity of intertemporal substitution
6 c o n s u m t r i a o t n e g r o w t-10 h 4 2 0 -5 -2 -4 -6 -8 -10 real interest rate 0 5 10 y = 0.2675x - 2.2871
The question that emerges is why would the elasticity of inter temporal substitution be so low in China? One possible factor is the financial position of the average Chinese as a net saver. When an increase in the interest rate happens, the positive effect on the income of this average Chinese is big enough as to overflow the temptation to save more and get a higher return. Also there is trend indicating that the elasticity of intertemporal substitution is rising:
Martin Aragoneses
Paper Draft
China Macro Anomalies
EIS y1990's = 0.3155x - 1.5162
20 0 -10 -5 -20 0 5 10
EIS 2000's y = 0.4085x - 3.1081
5 0 -5 -5 -10 0 5 10
A look back to the Historical Review
Macroeconomic Anomaly #2: Excess of physical capital accumulation
Investment (%GDP)
50 48 46 44 42 40 38 36 34 32 30 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Looking at the evolution of Chinese investment as a percentage of GDP over the last thirty year shows that investment rates are always over the 35% mark, going over
Martin Aragoneses
Paper Draft
China Macro Anomalies
40% at times, and after 2008 it has surpassed the 45% threshold (being the current estimates of the IMF 48.8% of GDP). These numbers are considerably high, even for a developing economy of the size of China. It is a question that we would like to ask is it is true that the Chinese economy is over investing. From our previous section, we know that anomalies in the sensitivity of the average Chinese to changes in the interest rate accompanied by an increase in the interest rates
Macroeconomic Anomaly #3: International Competitiveness During the period of transitional dynamics that China has experienced from being a de facto closed economy in the Maoist era to becoming the giant of global production that it is nowadays, some of the most interesting features of its economy developed. Next I will specify a model for the open economy starting with the well-known identity describing GDP as the aggregate of consumption by households, investment in physical capital by firms, government expenditure and exports minus imports: ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
Where we define net exports
We have studied the evolution of savings and investment in the previous two sections. Now we turn to the differences between the two. One of the key characteristics of the Chinese economy is that it has exhibited higher savings than investment in every year since 1994, this difference is known as the Current Account balance. Intuitively it means that, since the Chinese have an excess of savings with respect to the country’s investment needs, they are able to use this surplus to lend to countries with deficits (like the United States). Similarly, the current account surplus reflects the fact that, even
Martin Aragoneses
Paper Draft
China Macro Anomalies
though China has a considerable need for imports (e.g. oil), its exporting power is far greater. This is because we can express the current account balance as: ( ) Where ( ) ( ) ( )
reflects the fact that some Chinese have financial earnings abroad (such
us in the case of holdings of stocks) that should be included in the current account surplus, and also taking care of earnings of foreigners from assets inside of China (such as profits from an American company that has a production plant in Shenzhen). This variable also accounts for the remittances that Chinese workers abroad (for example the Chinese labor force building infrastructure in Angola) send back home to their families. We see right away in the graph below that the impact of the 2007/2008 financial crisis brought down net exports due to the especially in the year 2009, when Chinese exports took a major hit, having a negative growth rate in that year for the first time since 1990. Until now (in section 1) we have relaxed the definition of savings to simply the part of output that is not consumed. But “consumption” in section 1 referred to consumption by households only. Since now our model is expanded to include the part of output that goes to government expenditure and the earnings of Chinese national abroad minus the earnings of foreigners in Mainland China, we define national savings as: ( ) ( ) ( ) ( ) ( ) ( )
and thus we can derive that the current account balance of China is equal to the excess (or deficit) of national savings over investment. First we combine identities (A) and (B):
Martin Aragoneses
Paper Draft
China Macro Anomalies
Adding (C) to both sides and subtracting Y also to both sides we get:
Exchange rate WTO accession
Conclusions: Key Macro finding #1: when the interest rate increases in China, households’ consumption goes up and national savings fall, since the substitution effect is dominated by the income effect – the Chinese have a very low elasticity of intertemporal substitution ( ) due to the facts that Chinese are net savers to begin with, and the underlying inefficiency of Chinese financial markets. Key Macro Finding #2: are the Chinese over saving? Last assembly stage, low share of valued added. Problems in technology and innovation “new growth strategies”. Naughton During the financial crisis they start taking form, policy makers start intervening into the economy, promoting the surge of credit into the economy in 2009. The crisis was crucial because it gave the policy makers the incentive to shift the development strategy. New ambitious programs are started after the crisis to boost innovation. Naughton argues that the post crisis development strategy of Chinese policy makers is a “gamble”, the government still has the power to push China into a much more higher tech development trajectory. Can it work? Yes? With better educated labor force (as it indicates the surge in college graduates) and with the huge investment made by the government, at least some of this projects will work. But Naughton argues
Martin Aragoneses
Paper Draft
China Macro Anomalies
that the large commitment of resources to this development strategy will bring inflation due to the excess liquidity in the economy. High growth high inflationary trajectory.
A WORRYING DEVELOPMENT: One of the biggest problems that higher inflation will create in China is that real interest rates will be pushed down. This is because real interest rates are defined by well-known Fisher equation: ( Where is the real interest rate, rate. Thus )( ) ( )
is expected inflation, and is the nominal interest
Since both expected inflation and the real interest rate tend to be small, we can neglect the last term in the left hand side and write a good approximation of the real interest rate as:
This last equation tells us that, effectively, if Chinese policymakers are unable to control inflation, higher inflation expectations will depress real interest rates. From our analysis of the first macroeconomic anomaly we know that, due to the low elasticity of intertemporal substitution of the average Chinese (as well as looking back at historical data), a further fall in the real interest rate is likely to make aggregate consumption decrease, aggravating the problem of lack of internal demand.
Martin Aragoneses
Paper Draft
China Macro Anomalies
Inflation 25 20 15 10 5 0 -5 -10
Real Interest Rate
The State of the Chinese Economy: Implications for China and the World (2011) video
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 12 10 8 6 4 2 0 -2 -4 -6
Current Account Balance
Martin Aragoneses
Paper Draft
China Macro Anomalies
Household Savings, the Stock Market, and Economic Growth in China
Dayong Zhang and Yu Wu 2012
Martin Aragoneses
Paper Draft
China Macro Anomalies
r & savings (90-94)
45 44 43 42 41 40 39 38 -10 -5 0 5
r & savings (94-99)
44 43 42 41 40 39 38 37 -10 -5 0 5 10
r & savings (99-08)
55 50 45 40 35 30 -5 0 5 10
Martin Aragoneses TABLE 1: Real Interest Rate (%) Period 1 1990 1991 1992 1993 1994 Period 2 1994 1995 1996 1997 1998 1999 Period 3 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 3.3 1.7 0.4 -3.6 -8 -8 -1.5 3.4 7 7.3 7.2 7.2 3.7 3.7 4.7 2.6 -1.2 1.6 2.2 -0.1 -2.3
Paper Draft
China Macro Anomalies
National Svings, Gross (%GDP) 39.216 39.366 38.774 42.542 43.573 43.573 42.118 41.287 41.827 40.188 38.191 38.191 36.831 37.581 40.302 43.827 46.832 47.964 51.518 51.845 53.351
Consumtion by Households (%GDP) 46.72893216 45.38852989 45.74304001 42.662349 41.13192118 41.13192118 42.67962511 43.5099107 43.35147337 43.95590285 45.31081121 45.31081121 46.68679083 45.65513914 43.9727904 41.84702735 40.21603567 38.09693598 35.21472793 35.95624062 34.94356141
