China fact of the day

…the size of the Chinese government and party bureaucracy is surprisingly modest…In this respect, the Chinese communist Party is similar to previous Chinese dynasties as far back as the Han, which ruled the vast Chinese empire with a modestly sized civil service.

…China has only 31 government and party employees per thousand residents.  The number of civil servants per thousand residents in France is 95, in the United States, 75, and in Germany 53.

You will note that these numbers exclude state-owned enterprises, which in China are extensive although shrinking in relative terms.

That is from the new and excellent Nicholas Lardy book Markets Over Mao: The Rise of Private Business in China.  In my view the truth lies somewhere between the arguments of Lardy and the thesis of Joe Zhang, see the first Amazon review for Zhang’s critique of Lardy, plus Zhang’s comments here.  Here is Scott Sumner criticizing Zhang.


Zhang is right, but only in the rear view. The leadership is pushing forward a major reform of the state sector, centered on the banking/financial sector which is the only thing the inefficient state sector had going for it. Firms are being forced to divest their non-core assets, banks are under little pressure to lend them money, in fact the coal industry is being starved and steel may be next. Banks are also under pressure from money markets and they're going to need to raise hundreds of billions of dollars from the private market to cover their NPLs, so they will be brought under market regulation. The government role in SOEs is shrinking as well: one of the reforms ends all party jobs in management; any political control moves to the board room. In the past month, a private firm was given a license to import oil.

China is still a command and control type economy, albeit with price information from the outside, unlike the USSR. As for a few bureaucrats per 1000, the British Empire also had, supposedly, a mere room full of civil servants that controlled all of India (I think it was about 100 bureaucrats), but does anybody doubt Britain did not rule over India?

BTW I don't believe in the Economist graph that as of this year China SOEs assets are 45% of the economy, down from 66% in 2000, as reported in I think the numbers are either cooked, or the people running the supposedly private companies are members of the Red Army or Communist party.

If you think China is still a command and control economy you either don't know what a command and control economy is or what modern China is like. Though it does have its fair share of heavy handed regulations private businesses and customers are who determine the vast majority of resource allocation decisions. I recommend "How China Became Capitalist" by Coase and Ning Wang if you want a better understanding of the transition China has made. I have lived here for 4 years now, China is really different than what most people imagine sitting at home in America.

There is one important difference. Most of American life is outside the control of the bureaucrats. You can even joke about the DMV. All of Chinese life is within the power of that small number of officials. They have to fill out the things that go down on your permanent file which will follow you from place to place and job to job for the rest of your life.

You don't make jokes about the petty bureaucrats in China.

That waning but still present power over people's lives means if they want you to co-operate, you co-operate. They have no problems spying on your whole neighborhood. Getting reports on your boss. Nor are there any problems in having anyone thrown in jail for the rest of his life.

It is probably a mistake even to compare China with the West.

I actually think the polar opposite. If you have ever been to China you will see people doing all sorts of crazy things like starting a business without 15 government permits, cutting hair without a license and on and on. You never talk politics, but after that the typical Chinese enjoys a lot less harassment from government.

Yup. In the US, with a few exceptions, the government interferes much more in your actual everyday lives. The only compensation is a theoretical nanoslice of political power, in the form of the franchise. Feel that freedom!

It is the arbitrariness of the consequences of your actions in China that matters, but the action itself.

This is not necessarily a desirable state of affairs - see: absolute lack of trust in food safety. You can get almost anything in China these days, but the only thing I'm still requested to bring are vitamins / supplements / etc.

In any case, given how widespread internet usage is these days, I'd say the typical Chinese person is rather harassed by the govt every day - literally every time you go online, you're restricted by govt censorship in some form or another.

"Most of American life is outside the control of the bureaucrats."

It's still early but that might be the funniest thing I read today.

Do not, however, make the mistake of joking about the DMV inside a DMV facility.

Yeah, like harmless lemonade stands run by 10 year olds!

I wouldn't sign my name to a un-PC statement.

I assume this number does not include paid informants among the general population?

So, Germany is a socialist hellhole, with unions and strict green regulation, and has almost 50% fewer government employees than the U.S.

That cannot be the reason why Germany has such a large trade surplus, while the U.S. has such a notable trade deficit - must be the euro not being as strong as the dollar, right? (And for the American commenters - if the U.S. even cared about its remaining industrial base, there is no way in hell that the euro would be at its current exchange rate - but does anyone honestly think that the current situation is considered bad by the Greeks, Spanish, Italians, etc? That the U.S. is run by such inept policy is a god send to the countries that are still awaiting the eurogeddon predicted by American economists.)

Why do you mistake every example of German efficiency for 'socialism is great'?

I didn't read much of his comment, but I do like the idea of removing 22/1000 employees.

So what do you think it is? Please say "culture".

I'm sure part of it is culture.It sounds like you're rooting for culture. So am I. Understand that this is an unscientific mindset though.

According to the World Bank, still 35% of people in China is employed in agriculture. To keep the control of that system you need the military, not paper-pushers.

Indeed, 'control' also seems like something that becomes a labor intensive issue when you start getting urban populations and just greater population density in general. A gov't could infringe on a lot of freedom but need very few people to actually do the work if a good portion of the population lives in low density agricultral villages spread far apart.

Yes, I think perhaps, vast empires of mainly agrarian peasants tend not to need that much regulation, funnily enough.

It's when you get towns and merchants and oh my and all these people that have all these laws and rights and interests that the number of civil servants will boom.

Some private economic activity leads to booms in public sector activity, some doesn't.

What do the yankees do so much that needs people to think its important to have lots of bureaucrats around to manage it?

I would presume that, as a hierarchy, government size scales logarithmically with population.

When throwing up these sorts of statistics about the US, such as bureaucrats per thousand residents, please state whether you are talking about just federal civil servants, or including state and local bureaucrats as well.

Most comparisons of government in the US vs government in Europe that I see in the media compare federal (only) in the US vs national, local (and state in cases like Germany) for Europe and this sort of thing drives me up the wall. The effect is invariably to understate the size and reach of government in the US. There is an additional problem that alot of "government" in the US is conducted via supposedly private companies under contract, though with many of these companies obtain 90% of their revenue from government contracts.

I've seen the 100 British civil servants running India claim before.

First, I think the claim is a furphy. There was a government department in London dedicated solely to the Indian subcontinent. The British had naval bases in India, supported a large Indian Army linked to the British army, and constructed and ran an extensive naval network in India. The VIceroy alone probably had more than a hundred people employed in his personal household, let alone his administrative staff, plus there were subordinate Governors-Generals.

Technically, both the Indian Army and the Indian Civil Service were separate (though not really) from the normal British Army and British Civil Service, with their own methods of recruitment. I think at best the "roomful of civil servants" claim selectively counts some of the government employees involved in running India but not all of them, which is doable given that these employees fell into several categories.

There is also the issue that nineteenth and twentieth century European imperialism was done as much as possible through proxy, relying on existing native power structures. Even at the height of the Raj, a third of the population in the subcontinent lived in the princely states, where the method of rule was purely indirect.

I agree with your claims, and nobody is saying 100 English people made each and every decision in India, that would be impossible. Also your last paragraph is true, as I've heard the Mughal empire did essentially what the British did: rely on proxies. It's been said that a large portion of the taxes collected in India prior to the British (and probably I am guessing about 10% of GDP) when to just maintaining the Mughal palace(s).

Invaders like the British seldom have a problem co-opting the existing power structure to serve their own ends, even from a distance. See Vichy France. Bureaucrats don't care who they work for as long as they get paid.

East India Company?

The mere number of bureaucrats per thousand of residents is less important. It's their power which is all important. For centuries or millenia, Chinese bureaucrats formed an insurmountable obstacle to any progress. That's why the country was so backward in the early 1900's.

On the other hand, the number of civil servants in France is definitely overbloated.

Neither interpretation of this strikes me as good news. So, the Chinese get a lot of repression done with less. Yay.

Power corrupts, but absolute power is totally awesome!

Does this percentage include the number of contractors, you know the ones who acutally do the bureaucrats work?
At least that's true here in the US.

This is why it isn't useful to do comparisons of numbers of employees. Spending is ok.

I suspect the definition of "civil servants" might be different between countries. In China, the numbers are likely to exclude teachers, public transport, and government owned insurance companies. In the other countries, the numbers are likely to include all of these, thus skewing the ratio used for comparison.

"You will note that these numbers exclude state-owned enterprises, which in China are extensive although shrinking in relative terms."

Isn't that a huge, argument-nullifying exclusion (and props for pointing immediately to it Tyler)? Doesn't it basically make the following statement meaningless as a useful comparison?

"China has only 31 government and party employees per thousand residents. The number of civil servants per thousand residents in France is 95, in the United States, 75, and in Germany 53."

Yes it is. But it isn't stopping the confirmation bias in this thread.

As with accounting for government spending, accounting for government employment is largely an exercise in ignoring the obvious. The US figures do not, for example, include employment in the de facto nationalized health insurance industry. IRS rules if applied to the health insurance industry the way they are to private business would require the government to report them as employees and not independent contractors.

I'd imagine a lot of this hinges on how you count the few large pools of employees. For instance, the Federal govt in the US only has about 2.5 million employees total, and about 25% of those are postal workers. Is the post office privatized in other countries? Do they count military members?

Is having twice as many teachers (a plurality of state/local employees are teachers) evidence of more state control or just a richer society with a higher teacher:student ratio?

When the teacher's union is one of the nation's top five campaign donors and politicians vote for teacher pay and benefits, yeah.

Who are the other 4 out of 5 'top political donars'. Why are you afraid of political speech?

Measured crudely (nominal sums over a 25 year period), the top donors are a bundler for Democratic candidates ("ActBlue"), AFSCME, NEA, AT & T, and the National Association of Realtors. The top 3 give almost exclusively to Democrats, the fourth splits for Republicans 4/3, and the last splits evenly between the two parties. The 4th and 5th are in what have been abnormally regulated economic sectors (and appear to see politicians as fungible). The 2d and 3d are predatory, representing government employees wanting more privileges and more swag.

Yes, organizing by government employees incorporates conflicts of interest.

I'm sorry but who exactly would be a political donar who does not incorporate a conflict of interests?

Are you sure that AT&T and Realtors do not enjoy more swag and privileges from gov't policies?

"Is having twice as many teachers (a plurality of state/local employees are teachers) evidence of more state control or just a richer society with a higher teacher:student ratio? "

What does that really mean? Would having more teachers per student be similar to having more domestic servants per household? Economist Carl Menger was tutor to Hapsburg Crown Prince Rudolph, would that be the ultimate sign of a richer society? Does marginal utility apply to teachers? If half the population were teachers would we be better off in some way?

So what. This is a milder version of what Krugman said: that regulation does not impact the economy much because the salary of regulators is not very much compared to GDP.

A bullet does not weigh very much, but if one is shot in my head it will matter quite a bit.

All analogies are suspect, but this one is superb.

Krugman never said that. Try again

Without specifically stating the definition of government or party employee in each country, it is a useless comparison.

Lookimg again, I see that it says "government and party employees" when referring to China, and "civil servants" when referrimg to the Western countries. That already suggests there is an apples to oranges comparison going on.

US government employment is, in part, an employment guarantee program. It would be bad enough if they were all digging holes and filling them up again, but these hammers are looking for nails to justify their existence.

It makes me wonder how much of our growth is constrained by trapping all these resources into counterproductive efforts. Obviously some government employees are value enhancing in dealing with market failures. Too bad politicians dont know what a market failure is.

I'm not really following what you are talking about.

Basically US Federal employees are about 4M people and that number barely moves (and it's been moving down over the last few years). Back in the 80's it was 5M people. This, BTW, includes both military and non-military. The US population, in contrast, has grown from 226M in the 80's to nearly 320M today.

If anything on the Federal scale the US gov't is employing fewer people per thousand than it has in the past...that includes the military and the war on terror.

Yet as a 'jobs program', I'm not really seeing it. Leave aside the fact that 4M means only about 1% of people could have a job with the Fed. gov't. A 'jobs program' would be something designed to quickly add a lot of temp jobs during a recession and then ditch them afterwards. Most Federal jobs are either designed for people making a long career out of it (post office, FBI, career military etc.) or younger people doing a stint in the armed forces.

Ed debunked:

I’ve seen the 100 British civil servants running India claim before.,

But I think there's a germ of truth to this and in relation to Chiina. Both countries have huge populations. Whether the gov't is easy going and free or overbearing, it benefits from economics of scale. A single 'license raj' and related staff can handle nearly a billion people in India. In the old USSR the same type of office could only be over 300M people or so. So yes for good or bad very large countries are going to probably benefit from a civil servent to population ratio compared to smaller one.

I would suspect this metric might be more useful if you were comparing populations of similiar size. For example, Italy, the UK and France have almost the same population.

the number of civil servants in the US is an increasingly meaningless statistic. Particularly at the federal level, there are vast armies of contractors, several times the size of the civil service workforce, and it is official OPM policy NOT to keep track of their numbers.

I agree, but private industry often does this too. And it does lead to some serious accounting problems. There is a real difference between having a federal job or a job as a contractor for the federal gov't.

Likewise if we are going to start counting contractors how about counting private sector employment? What portion of, say, Boeing's employment, is due to getting government contracts? How about businesses that service Boeing's service of gov't contracts? How about people whose jobs are not directly tied to gov't work but are indirectly benefitted from them? Should real estate agents be partially included in gov't employment because some portion of their work is coming from indirect subsidies of homeownership by the gov't?

Boeing is mostly "private" - Lockheed was and Lockheed Martin is pretty much a for-profit government agency driven by profits to extract as much money from Congress as possible.

"You will note that these numbers exclude state-owned enterprises, which in China are extensive although shrinking in relative terms."

Smithfield Farms is operating according to the current five year plan; its acquisition was done as part of its new Chinese parent executing the five year plan for private Chinese enterprises in the food industry.

Milton Friedman was one of FDR's technocrats directing the US industry war effort. FDR did not need to, for example, nationalize the railroads, Wilson had demonstrated to industry that you either function in support of the nation at war or your assets will be seized and made to run by the government. Note Wilson merely followed the example of Lincoln who took over railroads which failed to serve the war needs. Both Lincoln and Wilson administration of railroads standardized the industry in critical respects which served the nation well after the assets were returned to shareholders in better condition than ever.

US industry once saw their future tied tightly to the future of the American people, because US industry would not have customers if the American people were not prospering. The Republican Party implemented big government like the ICC and anti-trust trust busting when US industry operated as if impoverishing the American people was a virtue.

Given corporations are held in much lower regard in the US by the American people than people in the rest of the world, corporations in China are clearly operating in the interests of China's people and thus don't need to be regulated like the Republicans did in their heyday when the US was much like China today.

It seems that the UH is about 84 - does anyone know where an international dataset for comparisons might be found?

Sorry I meant UK!

I am not sure if the numbers are right. US figure of 75 per 1000 residents means a total number of about 23.5 million civil servants.

Seems too high.

The One Touch Glucose Meter comes along with ttheir own diagnostic tests strips and also lancet.
This should fixx the problem of your Windows XP plug and play USB device
not working. There are actually windows made from several materials,
which include fiberglass, light weight aluminum, or vinyls are
a vriety of them.

Comments for this post are closed