China fact of the day

The richest 70 members of China’s legislature added more to their wealth last year than the combined net worth of all 535 members of the U.S. Congress, the president and his Cabinet, and the nine Supreme Court justices.The net worth of the 70 richest delegates in China’s National People’s Congress, which opens its annual session on March 5, rose to 565.8 billion yuan ($89.8 billion) in 2011, a gain of $11.5 billion from 2010, according to figures from the Hurun Report, which tracks the country’s wealthy. That compares to the $7.5 billion net worth of all 660 top officials in the three branches of the U.S. government.

The wealth gap between legislatures holds with statistically comparable samples. The richest 2 percent of the NPC — 60 people — had an average wealth of $1.44 billion per person. The richest 2 percent of Congress — 11 members — had an average wealth of $323 million.

The wealthiest member of the U.S. Congress is Representative Darrell Issa, the California Republican who had a maximum wealth of $700.9 million in 2010, according to the center. If he were in China’s NPC, he would be ranked 40th. Per capita income in China is about one-sixth the U.S. level when adjusted for differences in purchasing power.

That is via Shanghaiist, via @AlbertoNardelli via @AnnieLowrey.


Did being wealthy get them there, or being there made them wealthy?

Becoming wealthy is now one of the best ways to get yourself an invite to the Communist Party.

But don't feel too bad for the senior party members, who likely aren't part of the richest of the rich in the legislature. Having the power to make most of the important decisions for your economy and government spending behind closed doors without any accountablility has its privileges.

Exactly, the CCP offers NPC seats as bribes the the wealthiest people in each province, they also hand out seats to anyone famous enough to have any sort of personal following. It is a way of coopting & neutering them. Remember the National People's Congress could only generously be described as a rubber stamp body.

I'd also love to see the differences in wealth accrual in the five years after these public officials leave office.

Do they ever leave office? I don't know, I'm genuinely asking.

They do leave office, it is not like a life peerage, though being a member of the NPC is like being rewarded in the civil lists, it is a nice recognition of one's loyalty and patriotism.


Yow is right.

The question is, would China's markets be as free as they are if there weren't something in it for the politically powerful, and is the benefit worth the price?

The US budget is $3.6 trillion out of a GDP of $15.0 trillion. China's budget is $1.7 trillion out of a GDP of $11.3 trillion. Going by the amount of money taken from others, I expect US lawmakers still do more damage than the Chinese.

At some point China's growth will stop, and the legislature will become a rent-fest like that in the US. Until then, is it really worse to skim a smaller fraction of the growing GDP into their pockets than to put it into, say, ADM's pockets?

The Chinese State, which these communist leaders run, simply owns and/or controls about half of the major industrial/financial assets in the economy! Don't compare the U.S. budget-GDP ratio to the Chinese. It is better to compare the actions of U.S. lawmakers after the (partial) takeovers of GM, AIG, Fannie/Freddie to the normal state of affairs in China.

Yep, they're already rentseekers on a scale unimaginable here, which is why their growth is probably going to come screeching to a halt well before they are wealthier than Mexico on a per-capita basis -- assuming it hasn't already, their statistics aren't very reliable.

China's only a few years from matching (and then overtaking) Mexico on per capita income, so even if they experience a sudden downturn, I wouldn't expect them not to be at least the largest middle-income country in the world.

Mexico on the OTOH today seems like it is a only a few years away from a civil (drug) war......

Rahul: 'seems like' is correct. With the media's mantra of 'if it bleeds it leads', they make every place outside the first world seem like some violent Mad Max wasteland. I'd feel safer in Tijuana than downtown Detroit, for example.

Actually China's around $7800, Mexco is at about $14,400, so even assuming Mexico did not grow at all and China grew 8% a year, every year, it would still take them about 10 years.

If Mexico was growing at 3%, it would take 15, which is probably more realistic -- but I bet they collapse well before that.

It's pretty funny to realize that for all the talk of China's rise, they are still only half as functional as Mexico.


In 2010 Detroit had 34 homicides per 100,000 inhabitants. Tijuana statistics are harder to get but it seems the corresponding number is about 70. You might feel safer in Tijuana; but is it really?

It is possible that the Detroit murders were random, while the Tijuana murders were more predictable (say, only happening to active members of the drug armies).

In such case, an external tourist staying out of trouble would probably indeed be safer in Tijuana.

That said, unless msgkings is a speaker of the local dialect of Mexican Spanish, he probably feels safer than the place actually deserves. (Foreigner) ignorance may be a bliss.

China's government is also indirectly subsidized through currency controls. By not allowing the currency to float, they're essentially raking the delta between official and market value from Chinese companies who export.

That is an interesting statistic about the concentration of economic and political power. Or as a % of the respective country's GDP:

Richest 70 China lawmakers: 1.5% (=$89.8 bn/ $5.93 tr)
All 660 US lawmakers: 0.051% (=$7.5bn/ $14.59 tr).

You could add another order of magnitude by doing a "multiple of per capita GDP" calculation...

When members of the US congress discover this, they will move to increase their control of the military, police and news reporting organisations.

An interesting project might be to calculate the Gini coefficient for senates / congresses of various nations.

I wonder if this is also a characteristic of China that Thomas Friedman wants us to emulate.

I would say "occupy Beijing" but I think they tried that already.

Is that noise I hear Karl Marx spinning in his grave ?

No, that's Friedman laughing away.....

Is this good news for the rest of the world? (Making large-scale war less likely if their leaders are fat and happy, and with their personal interests increasingly tied to the global economy?)

A fair point.

Obviously one has to love the irony of a Communist country becoming a de facto oligarchy, but an average worth of a billion between them? I'm skeptical of the valuations, some of that has to be ghost cities and high-speed rail.

I am appalled. How have we let ourselves fall behind China in yet another measure of economic success?

For starters, we surely need to elect Romney. That will get our average up a little. Not enough, though. We need to convince Bloomberg to widen his horizons and join the national scene.

Best of all, though, it's time for Buffett to put his activism where his opinions are. Get into the Senate. That's sure to open a few eyes in China. Buffett, heed the call! One election can totally redress the wealth imbalance.

Having lived for some time in Hungary, I long felt that corruption--although morally reprehensible--was not the most important problem. Properly limited, corruption is only a transaction cost, no different than lawyers or investment banks (and potentially cheaper than the latter!).

Rather, the harm corruption causes is in dis-aligning social objectives from decision-maker objectives. So, the issue is not that a politician received 2% of the value of a bridge as a bribe for permitting it to be built; the issue is whether it is a bridge to nowhere or a bridge to somewhere. If it's a bridge to somewhere, then the bribe is merely an informal transaction fee. If it's a bridge to nowhere, then the entire capital cost and social benefit is lost. So it's not the bribe that matters, it's the bridge.

The value of the bribe is related to the value of the politician's position. (A shadow value, so to speak.) Chinese leadership is receiving "bonuses" of up to $100 million per leader. I'd like to legitimize this number, make it lower (due to a lower risk premium and Pareto Optimal behavior), and insure that politician incentives are aligned with social objectives.

In terms of economic or management theory, this is pretty plain vanilla. Considering the importance of the topic, I am amazed that no one discusses it.

But doesn't "corruption of the first kind" sooner or later devolve on to "corruption of the second kind"? Do we have any mechanisms or systems to limit corruption to the less damaging kind?

I would argue that the difference between Chinese and Russian corruption is that the Chinese have a business sensibility, the Russians don't. Thus, the Russians will demand bribes to so many people, that they cumulatively kill the project. The Chinese, by contrast, I think have a better sense of where "the market" is.

But, yes, corruption ordinarily is detrimental, because it leads to too many bridges to nowhere. Much of the history of, say, South America revolves around exactly this problem. And that is why any reasonable governance institution--and I mean you, IMF--should be explicitly incorporating principal incentive structuring into their mission work.

By the way, there's a valuation exercise here, in determining the value and nature of compensation to poiticians. "What's a Politician Worth?" might make a nice PhD thesis.

I'd be ok with each member of Congress getting a few million dollars a year if they could just keep from screwing everything else up. It would be a small price to pay.

I must concur with this opinion, being a Czech myself. Although the Hungarians are a different culture, the shared heritage of communism probably resulted in similarly aligned corruption patterns.

Looking at the price that Americans pay for the cancer-like swarms of lawyers everywhere: corruption per se is a smaller problem than overlawyering. Every society has its preferred way of rent-seeking.

Steven very correctly identifies "sustainable" and "malignant" corruption patterns.

This effect is likely only be due to the relative size of the two institutions (with 3000 vs 535 members) assuming a power law distribution of wealth.

I ran a 100 trials of Pareto distributed (identical distribution) lists of length 535 and 3000. The average of the total wealth of the top 2% over the 100 trials was about 6 times higher in the 3000 length case.

This would suggest that the China average wealth being only 4 times as much implies a much flatter wealth distribution in the Chinese governing body (or alternatively, a more unequal US Congress).

Always, always, always understand the distribution before making conclusions about the levels.

Wouldn't it be better to look at the average wealth of the top 2%, not the total wealth of the top 2%? The top 2% in china includes 6 times the members, so 6 times the wealth doesn't tell us much... Unless I misread you.

Maybe he needs to calculate the "average of the average wealth".

It ought to be size of sample independent, I think.

6 times the average is not all that likely (2% of the time), but 4 times the average is very likely (about 1 sigma away from the mean of 2 times as likely). The distribution of the ratio of the mean of the top 2% of one sample size of 3000 and one of 535 from the same distribution has a mean of ~ 1.9 and a standard deviation of ~ 1.4 after 100 trials. It is highly dependent on the distribution parameters. For "alpha" = 0.9 these numbers become ~ 3.6 and ~ 5.7 respectively.

Code (Mathematica):

DATA = Table[
RandomVariate[ParetoDistribution[1, 1.2]], {3000}]], -60]];
x2= Mean[
Table[RandomVariate[ParetoDistribution[1, 1.2]], {535}]], -11]];
x1/x2, {100}]

But it's the small sample sizes that does most of the heavy lifting. In heavy tailed distributions, small samples can have a large effect.

+1 to Jason, Ryan and Rahul for thinking like actual economists.

You are right; the large sample always gets a higher mean for the top 2%. For alpha=1.2 my result is identical to yours.

OTOH, the 100 trials does not seem sufficient to converge the ratio mean for the alpha=0.9 case. Tried a larger simulation; I'm getting a ratio of about 6 (but even at trials =4000 this doesn't seem converged)

So, is there a standard value of alpha for a Pareto wealth distribution in the Economics Literature?

My R code:
for(i in 1:trials)


There is a relationship between the Gini coefficient and alpha (q.v. the wikipedia page on the pareto distribution). However, Gini's vary a lot, and I think this relationship is based only on the tail of the distribution. Pareto distributions can be good descriptions of the high wealth tails of empirical distributions, but not for the low wealth side.

This paper derives a pareto distribution: it is actually ~ exp(-1/x^a)/x^a, but has correct tail behavior.

I'm not sure for alpha < 1 the results will converge since the mean does not exist (is infinite).

have to take a pay cut to have democracy? ... won't happen

Comments for this post are closed