What to Trust? Measuring the Chinese Economy

A question that is posed frequently by those skeptical over the health of China’s economy is: “If electricity consumption and rail freight traffic are both weak, how can GDP be expanding by more than 6%?” This is a great question because the answer highlights the dramatic pace of change in the structure of China’s economy. In today’s Sinology, we explore the reasons why the so-called “Li Keqiang Index” is a poor way to assess China’s growth, and offer some better metrics.

The structure of China’s economy has changed so much over the past decade that we need to change the way we measure its growth. Services and consumption have replaced manufacturing and construction as the largest part of the economy. These shifts have led, for example, to a decline in power consumption, but a sharp rise in express package delivery. In China, as in the U.S., growth is driven by consumer spending, and we have plenty of ways to check the official data. Apple, for example, reports that its greater China revenue rose 14% in the most recent quarter, and Mercedes sales were up 50% in January. And, it is worth noting that power consumption by the services and consumer-related sectors rose 7.5% in 2015.

It Began with a 2007 Dinner Chat

The origin of the focus on power and rail freight comes from an obscure source—a private dinner conversation in early 2007 between then U.S. Ambassador to China Clark T. (Sandy) Randt, Jr. and current Premier Li Keqiang, who at the time was the Communist Party Secretary of Liaoning Province.

Li Keqiang was invited to the ambassador’s residence in Beijing because he was a rising star within the Party leadership. (The following year he was promoted to vice premier, and in 2013, he became premier.) When Ambassador Randt asked Li to comment on economic conditions in 2007, Li’s response, according to official embassy notes, was: “GDP figures are ‘man-made’ and therefore unreliable.”

The notes also indicated: “When evaluating Liaoning's economy, [Li said he]focuses on three figures: 1) electricity consumption, which was up 10 percent in Liaoning last year; 2) volume of rail cargo, which is fairly accurate because fees are charged for each unit of weight; and 3) amount of loans disbursed, which also tends to be accurate given the interest fees charged. By looking at these three figures, Li said he can measure with relative accuracy the speed of economic growth. All other figures, especially GDP statistics, are ‘for reference only,’ he said smiling.”

The ambassador’s report on his conversation with Li was classified as “CONFIDENTIAL,” meaning it was only to be read by U.S. government officials with security clearances. Then came WikiLeaks, which made this document, in addition to thousands of other classified reports, available to the public, and that led some analysts and pundits to refer to the “Li Keqiang Index” for the Chinese economy.

Superficially, this index sounds sensible, especially since it came from a senior Party official who thought he was speaking privately. More careful scrutiny, however, reveals Mr. Li’s index to be as backward-looking as judging the health of the American economy based on sales of BlackBerry phones . . . or the growth of power consumption.

Power Problems

Let’s start with electricity. Significantly slower growth in power consumption reflects the rebalancing of China’s economy, away from heavy industry and toward services and consumption.

In 2006, the year before Li spoke with the ambassador, China’s construction boom was nearing its peak, and the economy was driven by power-intensive industries, such as steel and cement. Total industrial value-added rose 17% year-over-year (YoY), steel production rose 18%, cement output was up 16%, and the secondary part of the economy (manufacturing and construction) accounted for 47% of GDP. Liaoning, where Li was the top provincial official, had a very high concentration of construction-related sectors, and in his dinner with Ambassador Randt, he boasted that provincial GDP growth was 12.8%.

Nationally, booming heavy industry drove electricity consumption up 14% in 2006, while GDP growth was almost 13%. At that time, across China—but especially Liaoning Province—there was a very high correlation between power and economic growth. But, the times they are a-changin’.

The peak in the growth rate of China’s fixed-asset investment (FAI)—a good indicator of construction activity—was 30% YoY in 2009, but that slowed last year to 10%. Industrial value-added last year rose 6%, down from 17% in 2006. Steel production declined 2% last year, compared to an increase of 18% in 2006. As a result, last year, China’s overall electricity consumption rose only 0.5%, and industrial power consumption was down by about 1%. But, does this signal that GDP growth should have been flat or negative?

Last year, the tertiary (services and consumption) part of the economy accounted for just over 50% of GDP for the first time, compared to a 42% share in 2006, while the secondary (manufacturing and construction) share declined to 41% from 47% . The power-intensive part of the economy shrank while the less energy-guzzling services sector expanded. Last year, consumption contributed 66% of GDP growth, compared to a 42% share in 2006.

Industry consumes 71% of China’s power, but is a shrinking part of China’s economy, so it isn’t surprising that overall power use is growing much more slowly than the overall economy. On the other hand, power consumption by the services and consumer-related sectors rose 7.5% last year, and residential power use was up 5%.

The structure of China’s economy has changed significantly over just a decade, rendering electricity consumption a far less meaningful gauge of overall economic activity. Consider that in the U.S. last year, overall power consumption fell 1% YoY, with industrial use down 4%, while GDP rose by more than 2%. Like China, the U.S. economy is driven more by services and consumption, and less by power-intensive heavy industry, weakening the correlation between the growth rates of power consumption and GDP.

Finally, I want to mention one other aspect of the power issue. Some have argued that electricity consumption is one of the few Chinese data sets they trust. But, if you think Beijing is lying about most economic data, why do you trust the power numbers? Almost all of the power is generated, transmitted and distributed by state-owned firms, so if the government is attempting to cover up a much steeper economic slowdown, why aren’t they fudging the power numbers as part of that effort?

All the Live-Long Day

The second component of the so-called Li Keqiang Index is rail freight. Back in 2006, the volume of rail freight rose by about 7% YoY, while last year it declined 12%. As with the power story, this reflects significant structural changes.

The biggest change is that in 2006, rail carried 14% of China’s total freight traffic, a share that fell to only 7% last year. Highways now carry 79% of total freight now, up from 72% in 2006, and highway traffic rose 6% last year. Focusing solely on rail traffic today would be akin to tracking sales of print media in the U.S. while ignoring online readers.

It is also important to note that a key reason for significantly slower growth in rail freight traffic is that more than 50% of rail freight is coal. Thus, to a great extent, slower growth in rail traffic reflects slower growth in power generation, which, as I explained earlier, reflects slower growth in energy-intensive industrial sectors.

And, consistent with the strong growth in the consumer economy, especially online retail, express parcel deliveries in China rose 48% YoY last year, after rising 52% in 2014.

Bank Loans

The final component of the so-called Li Keqiang Index, according to the U.S. embassy report, is the amount of loans disbursed. This was an odd choice for Li, given that all of the country’s banks are state-controlled and aggregate lending is limited by a government-set quota. But let’s take a look at what these numbers tell us.

New bank loans disbursed last year were about 2.7 times greater than the amount issued in 2006, and the average annual increase in new lending was 17.7% during that period.

This is not, however, our preferred metric for monitoring the availability of credit. Aggregate credit—including bonds, entrusted loans and other financing channels, plus bank loans—is a better gauge. It is something China’s central bank calls Total Social Finance (TSF). Moreover, I prefer to monitor the growth rate of TSF outstanding, rather than just the new flow, as that is a better reflection of the availability of credit.

TSF outstanding rose 19% YoY in 2006 and then 22% in 2007, before rocketing to 35% in 2009 in support of China’s huge stimulus response to the Global Financial Crisis. By 2011, the growth rate of TSF outstanding was back to the 2006 level at 19%. Last year, the growth rate was 12%.

As the next chart illustrates, the growth rate of aggregate credit outstanding has decelerated in line with the deceleration in nominal GDP growth—a sensible change, in my view.

What Does Work?

If not the Li Keqiang Index, then what metrics should investors use to assess the health of the Chinese economy? For starters, not the GDP growth rate.

GDP growth may be the least important statistic in China. After all, we don’t make investment decisions in the U.S. or Europe based on GDP growth rates. As in other markets, the important stats in China concern employment, income, inflation and consumer spending. In my view, we have far better visibility into these topics now than we had in the past because of the growing role of privately owned Chinese firms, the significant participation by foreign-owned firms in the domestic economy, and because foreigners are now able to conduct independent surveys of business conditions.

China’s old economy was weak last year, and that will continue this year, especially in heavy industries, such as steel and cement. China has passed its peak in the growth of construction of infrastructure and new homes. But manufacturing has not collapsed, with a private survey revealing that wages at privately owned factories were up 5% to 6% in 4Q15 (compared to 6% to 8% a year earlier), reflecting a fairly tight labor market. Crude oil imports by volume rose 9% (after rising 9% in 2014), and the volume of copper imports was up 12%.

I expect the consumer story to remain strong for several reasons. Income, while decelerating along with the rest of the economy, should grow by more than 6% in real terms this year. This follows a decade of more than 130% real income growth, compared to about 11% growth in real per capita disposable personal income in the U.S. over the same period. I expect another year of moderate consumer price inflation, at about 1.5%. Household debt is very low, and the savings rate very high. All of that contributed to an 11% rise in real retail sales last year.

We can check that official retail sales statistic against numbers from private sources. Apple, for example, reported that during its 1Q16 period its greater China revenue rose 14% (compared to a 4% decline in the U.S.) while Nike said its shoe sales were up 30% during the quarter ending in November. The Japanese government told us that last year, Chinese visitor arrivals rose 107%, after an 83% increase in 2014, and that Chinese spent more than twice as much as the average visitor. In January, Ford’s vehicle sales in China rose 36% YoY, while Cadillac sales increased 16%, the sixth consecutive month of double-digit sales growth, according to GM. Mercedes reports that its unit sales in China rose more than 50% in January.

Slower, But No Signs of Collapse

As I’ve been saying for some time in my commentaries, it is inevitable that most Chinese economic statistics will grow at gradually slower year-over-year rates for many years to come. This is due in part to structural changes, including a shrinking workforce, and because after three decades of 10% growth, the base has become too big to sustain double-digit expansion.

Too few investors (as well as too few pundits and journalists) recognize that 2015 was the fourth consecutive year in which the manufacturing and construction part of the economy was smaller than the consumption and services part. Consumption accounted for about two-thirds of GDP growth, illustrating that the economy has rebalanced away from a dependence on exports, heavy industry and investment, and has, in my view, become the world’s best consumption story. This mitigates weakness in manufacturing and construction, and, if this rebalancing continues, it should mean that macro deceleration will be gradual.

Andy Rothman
Investment Strategist
Matthews Asia

As of February 29, 2016, accounts managed by Matthews Asia did not hold positions in Apple, Inc., Daimler AG, which owns Mercedes, BlackBerry Limited, Nike, Inc., Ford Motor Company and General Motors.