It is clear from the comment section to the previous post that there is a lot of confusion about PPP. Let me try to illustrate the basic ideas with an example, since my words don’t seem to be making an impact.
Assume that Saudi Arabia has a nominal GDP of $1000 billion, of which $600 billion is non-oil output and $400 billion is oil output. Let’s assume that the domestic and international prices of oil are the same for Saudi Arabia, but that other goods are 3.5 times as expensive in the US as in Saudi Arabia. In that case the Saudi GDP in PPP terms would be:
3.5 X $600b + $400b = $2500b
You might think about there being a “multiplier” of 3.5 for non-oil output, and a multiplier of 1.0 for oil. The overall multiplier is 2.5 = $2500b/$1000b.
Now suppose you want to find the non-oil GDP in PPP terms (which I assumed was $2100b), how do you do that?
You started with the total output (from the IMF or World Bank) in PPP terms, which is assumed to be $2500b. Then you subtract the output from the oil sector, using international prices. In this case I assumed the domestic and international prices were identical. Now suppose the domestic price of oil in Saudi Arabia were less than the international price. In that case you’d want to subtract out oil at the international price, not the domestic price. Do all PPP prices or all nominal prices; don’t mix the two.
Some commenters wanted to subtract the oil sector from nominal Saudi GDP. But I’m not interested in nominal Saudi non-oil output; I want PPP non-oil output. And if you apply the overall 2.5 multiplier to the non-oil sector, you get the wrong answer, as the actual multiplier is 3.5 for the non-oil sector (by assumption), and 2.5 for the total economy.
This is from a recent article in the Economist:
Cheap oil is forcing Gulf monarchs, who have hitherto bought their people’s acquiescence with cushy jobs and handouts, to trim the public payroll. And since Gulf monarchies cannot find enough jobs for their own people, the safety valve of emigration to work in the Gulf has closed to other Arabs. The largest Gulf state, Saudi Arabia, needs to create about 226,000 new jobs every year, according to Jadwa Investments, a Saudi research firm. But in 2015 employment rose only by 49,000.
Gulf states have set quotas for the employment of nationals, but many companies complain that local graduates lack the skills and work ethic required. “I know of firms that pay Saudis to satisfy the law, but tell them to stay at home,” says one businessman.
Now of course the foreign workers in Saudi Arabia are a different story. But many of them are low skilled construction workers, maids, etc. So I still find it kind of amazing that Saudi resident workers seem to be more productive than German workers, even if you exclude the entire oil production industry from the Saudi data. But that’s just me.
Or maybe (more likely) the IMF/World Bank/CIA PPP estimates are all inaccurate.
Also note that having lots of money doesn’t magically solve productivity problems. Countries in the “middle income trap” are often saving enough so that they ought to be catching up to the developed world. The fact that they are not doing so suggests that productivity problems are much deeper than a lack of money.
If I had to argue against my anti-Saudi prejudice, here’s what I’d say:
Some places are unsentimental and very smart. They realize that they can have living standards far above New York City through international labor arbitrage. They understand that NYC does not have “nice things” like good subways and good roads and good airports, because they insist on using union workers and local firms that are not skilled at building subways. So the subways cost three times more than even in Europe. These unsentimental smart places like Singapore and Dubai realize that they can bring in the best foreign firms and very cheap construction labor from Asia, and build great infrastructure for low prices. If you then price the output of the infrastructure at America prices, it looks really impressive. So maybe the Saudi’s really are highly productive, because they do this sort of labor arbitrage. They sell oil at American prices, and hire Bangladeshi construction workers at 5% of American wages. Sounds good to me!
When I read Dems (and Trump) talk about the need to build infrastructure in order to create “jobs”, I’m reminded of Milton Friedman’s famous joke. If you want jobs, use spoons. If you want NYC to have nice infrastructure, use Bangladeshi workers.
PS. I just saw this, from the WSJ:
The Wall Street Journal reached out to 45 economists who have served on the White House Council of Economic Advisers, under both Republican and Democratic presidents, to ask about this year’s presidential election. Most Democratic appointees said they supported Hillary Clinton, while no Republican appointees openly supported Donald Trump.
I know, the experts have “screwed things up” so why not give outsiders an opportunity? Yes, instead of being screwed up as we are in America, we can have a policy-making apparatus run by non-experts, and get screwed up in the way that non-expert dominated places are screwed up, as in North Korea, or Venezuela, or Zimbabwe.
When you are a country that is already in the 97 to 99 percentile of the global economy, in terms of overall economic success, I’m not sure you want to just kick out all the experts and try something completely new. Your upside if it works might be Singapore or Switzerland. But your downside risk . . . ?
I wrote this back in 2011:
Now let’s start down through Dante’s seven circles of Hell:
1. The US is much richer than Mexico. So much so that millions of Mexicans will risk the horrors of human trafficking into the US to get crummy jobs picking tomatoes all day in the hot sun.
2. China in 2011 is still considerably poorer than Mexico. The Chinese take much greater risks to get here.
3. China today is so much richer than China in 1997 that it’s like a different planet. The changes (even in rural areas) are massive.
4. The China of 1997 seemed like paradise compared to the China of the 1970s. Throughout Hessler’s book, people keep talking about how horrible things were during that decade and how prosperous they are now (1997 in Sichuan!)
5. The China of the 1970s was nowhere near as bad as during 1959-61, when 30 million starved to death.
It’s a loooooong way down to the lower percentiles.
PPS. Yes, I know, there were nine circles. I have a bad memory.