I went for a seminar last week about the reasons for the very high Chinese savings rates, which has been blamed for the cheap capital flowing into the US during the housing crisis and is thus the root of all evil. It was presented by Juann H. Hung, a Taiwanese who worked at the Congressional Budget Office at the Macroeconomic Analysis Division at the time of writing, and the paper was also written in cooperation with the World Bank. The research considers data from 1980 to 2006.
It pretty much dispells a lot of notions of “why” this crisis happened. Like any friend of history would have told Mitt Romney – the Plaza Accord didn’t help the US with its uncompetitive exports and only served to send Japan over the cliff and the US into an economic bump – Mrs. Hung’s analysis states that the Chinese currency is undervalued but it doesn’t do much to impact the high savings rate.
Summary and some of my comments on the individual factors contributing to the Chinese savings rate:
- Low proportion of old people – People are young enough to earn more than they spend! Also younger than needing to invest for more people than themselves.
- Strong growth rate – Because the working population is the earning population, they are disproportionately able to save as the economy grows relative to younger or older people who contribute into the economy only as consumers.
- Weak social safety net – Sick? Not Studious? Dismembered when working? Tough luck, better hope you managed to save from whatever means of earnings you used to have. Insurance and pension schemes are nearly unheard of in China, and property is king.
- Low urbanization – farmers have to take a lot more risks than city-dwellers. Will it be sunny enough this summer to grow? Will there be enough rain for all the crops to just not whither? Will I be able to sell my produce at a decent price if other farmers had a much better crop in the rest of the country? City-dwellers get paid by the hour of work, not on the hour of sunshine, and thus need considerably less savings. Considering that China is around 40% urbanized, this is a major effect!
- Being a low income country – Largely captured by the above factors, but lets make it clear: people in low income countries simply save more! The effect however isn’t very elastic with Chinese data – I will digress on this later.
- Being an East Asian country – Japan, Taiwan, Singapore, Hong Kong, South Korea. Is the first term that comes to mind “personal indebtedness”? All have high savings rates in the population from Confucian values, and most of them have embarked on growth paths that encourage personal savings via financial repression to maintain cheap capital to the conglomerates.
- Currency undervaluation – yes, the currency is undervalued, and there is some data on this after the list. Horrible, but how much does this effect savings? Less than 1%. Currency revaluation will not stop the flow of cash exports out of China!
- Real Interest Rates – increases will effect the economy via increased attractiveness of saving, while also impacting savings negatively on a proportional (not absolute value) basis. Currently, if the interest rates go up, the savings goes down in a proportional relationship of about 15% meaning that if interest rates double, savings every month decrease by 15%. A part of this is the inflation adjustment, and inflation has a positive effect on savings, mostly from being seen as an uncertainty barometer in underdeveloped countries.
- Scarcity of domestic credit – the thesis is that if you don’t have access to loans, you better save to make investments in business or anything else. Supported, but there is only a 3% effect in China on proportional changes.
- Having undergone the Asian Financial Crisis recently enough to have it in clear memory – this is another important factor, although it has a lot of variability depending on whether you include specific countries. The lesson China drew from this crisis were simply deep enough that they can now hold the lectures!
There is a huge problem with the data set, acknowledged by Mrs. Hung herself: it does not compare the East Asian growth periods of Korea, Japan Hong Kong and Singapore as well as Taiwan (partially for the latter) when they grew from a base comparable to where China started, or even where it is today. Getting homogenous data is obviously very difficult, but it does dilute the value of the East Asian dummy comparison used to absorb unquantifiable factors.
Also, the paper does not address developments after 2006 for statistical purposes, but does mention some reasons why the Chinese savings rate post-2001 has been increasing steadily:
- Point 10 above is included here as well, because it could actually be scored and measured using the model employed by Mrs. Hung and it is showing a very high importance.
- China’s one-child policy. Huh? What does this have to do with savings? It’s a long train of thought, but it starts in the farmer village. Daughters will be married into the families of their respective husbands, and the individual couple will then be caretakers of the husband’s family as he is the breadwinner for his parents during retirement. Thus the husband’s family has support, and the daughter’s do not. This leads to the daughter’s family needing to save more for old age or simply not (officially) have daughters. China thus has a history of “miscarriages” and there is a surplus of roughly 20 million young men! This has another interesting effect – women can pick and choose whoever they like. One important measure for any man is of course financing, more specifically his ability to buy a house for the family. Renting an apartment is simply not going to cut it. Besides the bride’s family needing to save their family wealth in old age, the groom’s ditto need to save up huge amounts now just for him to court her and secure a marriage! This has been an increasingly important effect since 2002, and evidenced by authors Wei and Zhang in a 2009 paper.
- Michael Pettis argued in a 2010 paper that the state bailouts of the banks’ nonperforming loans portfolio from 2001 onwards is highly responsible. He blames three methods: First, the central bank discouraged new loans by keeping lending rates and thus cash inflow low, allowing banks to work through their bad loans portfolio on the back of economic growth. In conjunction with this, the second effect of keeping borrowing rates high created a very big interest rate gap which fueled bank profitability Third, the Chinese government invested heavily in bank equity and bad loans at above-market rates to allow balance sheet cleansing. Astute readers might now ask me to hold my horses, as well as: do not all these things decrease savings? Well, on the personal level, yes, but it allows overall savings nationally – considering corporate saving as well – to rise. Mostly, the level of personal income is decreasing as a share of GDP, and corporate cash grows.
- One personal expectation is that internal migration in China is largely responsible. People moving from the poor inland regions in China to Tier 3 or better cities have a bigger incentive to save than the average Chinese as it gives their relatives back home a bigger payout in PPP adjusted terms. Additionally, it provides the promise of a better retirement for the saving individual while mitigating all wealth effects of a higher salary. This group of people is vast in China.
Returning to point 5 – Being a low income country, and its importance despite the increases in Chinese income during the 26 years measured not being evidenced by correlation data, it has a fairly simple explanation. The model used is a system of two equations, where the first is the level of all factors, and the second system of equations is the annual change in the factors of the first system. The most important single term in virtually all regression runs was the previous period of savings, included because saving is either sticky (by direct deposits of salaries into a savings account or built by habit) or dependent on longer-term trends than can be captured in annual changes. Thus, prior savings was the most important factor, and it makes intuitive sense – if you have been raised with money being scarce as the Chinese productive class has been, you won’t change the impression just because growth is strong – savings- and spending patterns are a huge part of one’s personality.
To continue the currency discussion in point 7: in terms of productivity (used in this paper) it is 47% undervalued! Whoop dee doo! Hold your horses though: the measure used is productivity, or how much the country on average exchanges in one unit of output. Extending this line of thinking, it is also indicated in the paper that the OECD average is overvalued by 55%! If you are reading this, chances are that roughly 40% of what you are earning (1 – 1 / 1.55) is simply from being in a developed country, whereas the Chinese worker is earning half of what they should! Again, how much does this effect savings? Less than 1%. Currency revaluation will not stop the flow of cash exports out of China!
There was very little discussion of “what do we do about this, if it is really a problem”. Still, there are lessons here, although I feel that they may be justified to bring into a later post.