What is the Wealth of a Nation?
aka GDP is a lie and the price of fuel and eggs is more important
I’ve been pondering the problems with GDP statistics for a while now. Economists love them, and they do give an idea of how wealthy a nation is, but rather like the accursed BMI of obesity studies, they are also highly misleading. In fact, just as BMI was useful on a population level when it wasn’t being closely monitored, much the same applies to GDP. With the disadvantage for GDP that economists, bankers, traders and ministers of finance have been tracking GDP closely for a lot longer and have learned to game the system.
GDP - What is it?
The usual way people calculate GDP is by spending. That means simplistically the GDP of a country is the sum of consumer spending, business spending, government spending in the country plus net exports (i.e. exports - imports). It is also in theory possible to calculate it by income and the numbers should be the same, but the income calculations are typically considered to be harder because of required adjustments.
But GDP, even as defined, has some problems.
First of all, and this may come as shock, some countries lie about their GDP and related statistics. And by lie I mean flat out falsify them in the ministry of finance or whatever other government body is charged with calculating them. West Taiwan is an example of this, so is Russia, but they aren’t alone in this trickery. Other countries may not intentionally lie, but end up producing misleading statistics because the data sources they use to calculate them are erroneous and are not cross checked. For example countries with a large informal gray market sector have to estimate how much that sector contributes because these transactions, being by definition illicit, are not accurately tracked by the government or, generally, any other institutional body. Mixing and matching knowingly lying and the gray market, IIRC, Italy (or Greece or both?) at one point in the 1990s suddenly recalculated the amount of GDP in the gray sector and therefore, magically, hit the required economic targets to join the Euro.
And then, even among nations that have good, honest statisticians and data gatherers, there can be disagreement about how to account for some statistics such as how much weight to give to various sorts of government spending. For example five years ago during the Covidiocy the UK’s statisticians (correctly, IMO, and also I believe uniquely) cut the GDP impact of state spending on education because they said that remote learning was ineffective and thus the government spending was just wasted. The result of that was that the UK’s GDP took a bigger hit in 2020/21 than other similar nations and then in 2022/23 the UK had greater GDP growth. Another example is Russia since 2022 which has reported GDP that is (when not fictional) almost entirely due to government war spending that has been remarkably ineffective.
But yet lots and lots of people treat GDP as a gold standard metric set in stone that is truly perfect. They also treat derived metrics such as GDP per capita and the definition of recession being two quarters where GDP declines as the same. Sadly they are misguided because GDP has fundamental flaws. Here are two of them.
GDP and Government
I’m not alone in thinking that government spending needs heavy adjustment before it gets included in GDP and possibly should be entirely removed from it. The AP (putting a good anti-Trump/Musk spin on things) reports:
Commerce Secretary Howard Lutnick said Sunday that government spending could be separated from gross domestic product reports, in response to questions about whether the spending cuts pushed by Elon Musk’s Department of Government Efficiency could possibly cause an economic downturn.
“You know that governments historically have messed with GDP,” Lutnick said on Fox News Channel’s “Sunday Morning Futures.” “They count government spending as part of GDP. So I’m going to separate those two and make it transparent.”
Doing so could potentially complicate or distort a fundamental measure of the U.S. economy’s health. Government spending is traditionally included in the GDP because changes in taxes, spending, deficits and regulations by the government can impact the path of overall growth.
Note the bolded (by me) clause. Regulations frequently put the brakes on economic growth, but regulations are implemented and enforced by regulators - that is to say government employees. So if you have to hire an additional 10,000 EPA regulators to enforce forbidding the development of something useful, that could potentially increase GDP while reducing actual wealth for citizens.
An Axios article on the same topic makes a similar point:
Zoom in: It is true, however, that government spending is counted in GDP by simply adding in the dollars spent, without any real test of how efficiently or productively the money was used.
"If the government buys a tank, that's GDP," Lutnick said in the TV appearance. "But paying 1,000 people to think about buying a tank is not GDP. That is wasted inefficiency, wasted money. And cutting that, while it shows in GDP, we're going to get rid of that."
It's true, as Musk and Lutnick suggest, that if the government hired a bunch of people to twiddle their thumbs all day, it would show up as higher GDP while not making anyone better off, save perhaps the thumb-twiddlers.
Now it is true that economists/statisticians often remove some government spending from their calculations of GDP and what they remove can vary between different people (see UK example above) but the regulation example I make above shows that this may not be enough. Removing regulators and regulations may increase growth both directly by allowing more things to happen and indirectly by making the speed of bringing new things to market faster with lower capital requirements because (for example) the previously required 300 page environmental impact survey that needed a team of a dozen to create over six months is now a one pager that one project manager can write in an afternoon.
GDP and Technology
But government spending is not the only issue with GDP. GDP has problems accounting for increases in quality and in increases in leisure time. Both of these make people feel richer but they generally aren’t changes in expenditure.
For example if I replace my several years old computer or smartphone with a new one I’ll probably get a device that has more memory, more CPU, a better screen and so on for the same price as the old one. I am wealthier. But GDP doesn’t account for the difference. If the new devices are cheaper than the old ones it may even look like it’s a reduction in GDP, which is nuts.
It gets worse. Because of how GDP is calculated when you spend money on services like laundry, cleaning, gardening or child-care (or dining out) those add to GDP. When you do those things yourself (or your spouse does) then it does not. Hence a washing machine, which Hans Rosling discussed the revolutionariness of in this Ted Talk, potentially causes a reduction in GDP because a family no longer pays a washerwoman to do the laundry even as it makes the family cleaner (and hence presumably healthier) because now they can do laundry every day or two instead of once a week.
Arnold Kling writes more about this in some of the following substack about AI but AI is just the latest example following the Internet, Personal Computers and so on.
I made a comment there that is worth repeating.
If I can use an AI to write half a dozen emails that would normally take me 10 minutes each that's an hour of my life I've got back (less time taken to read the emails before hitting send). That number totally fails to show up in GDP
Worse, if your case 7 applies and we can use AI to program our lawnmowers, roombas etc. we'll see a diminution in the amount of human provided services in those areas and thus, for those people who pay for gardeners etc. rather than do it themselves, a reduction in calculated GDP because they no longer need to pay Jose, Maria and their friends to do it for them. (Although some of those services are already missed by GDP since they are paid in cash and undeclared by the recipient). But that dimunution is in fact false because the people who used to hire gardeners and cleaners and now have a robot to do it are saving time and money thanks to the new technology. That ought to make them, and the nation, richer not poorer
A general rule though is that technology improves what we can do individually and thus cuts out middlemen. That tends to look like a negative for GDP because the middlemen lose their income and jobs and hence their spending power.
Other Measures of Wealth
As Kling responded to my comment on his substack, (free) time is a pretty good partial measure of wealth. If you don’t have to work 16 hours a day to feed yourself and keep a roof over your head then you are in many respects wealthier than someone who does. And the same goes to some extent for people who don’t work at all compared to those that do. Though few would see people living in shitty social housing on a bare minimum social security benefit as richer than big city lawyers pulling down telephone salaries and working 70 hour weeks to live in a big house in a leafy suburb, so at some point leisure time is clearly not a sufficient measure of wealth or prosperity.
If you have freedom of choice about what to buy, what to do etc. you are wealthier in many respects to those that don’t. That’s another thing that GDP fails to capture. A country where you spend $20 at a restaurant and your only choice is a meal of cabbage, potatoes and sausages looks, to an economist, much the same as one where that $20 can get you your choice of fried chicken, curry, noodles, hamburgers, fish and so on. If you can buy your choice of food at the time of your convenience that’s also a sign of wealth. Can I get my meal of fried chicken at 9pm? how about 3am? or is it only available 11am-2pm and 5pm-8pm? None of this goes into GDP or into a bunch of other statistics but a society where anyone can get a $20 meal of their choice of fried chicken, noodles or hamburgers at 3am is clearly richer than one where you can only get cabbage, potatoes and sausages as a meal between 11am-2pm and 5pm-8pm.
And that leads to a separate, but hopefully obvious point. Individual prosperity is different to national prosperity. And GDP per capita is a poor way to identify individual prosperity (wealth). However generally speaking its better to live in a richer country than a poorer one based on GDP or similar metrics.
Your prosperity correlates with but is not identical to that of your country
To a degree this should be obvious. There are very rich people in poor countries (see e.g. Putin and any number of other oligarchal/dictatorial sorts) and there are poor people in rich countries (see the homeless in the US etc.), but on the whole national prosperity correlates with individual prosperity. Poor people in rich countries often don’t realize that in certain ways they are richer than rich people in poor countries. I made that point some years ago discussing the Soviet Union:
There’s a famous set of pictures of Yeltsin looking amazed at the contents of a US supermarket and I can confirm that the absolute same occurred to Soviet students that I showed around the UK in December 1989. […] When we took them to London they were unimpressed with the Houses or Parliament (which we’d worked hard to get a tour of) but utterly blown away by the fact that we could stop on the way back on a random street in East London for a late snack and have a choice of curry, kebab or fish and chips from three different take aways near by[…] Oh and at the same stop we also stocked up on booze and cigarettes at the off-license corner shop across the street. […I]t absolutely astonished the Soviets who marveled at the selection available. There were half a dozen types of vodka, more whiskies, various gins, rums etc. plus multiple brands of beer, wine from all over the world (well France, Italy, maybe Germany, and Australia) and perhaps 20 brands of cigarettes and cheap cigars. To the average Western European, Australian or North American this sort of selection was normal but it was inconceivable to a Soviet. Let’s face it a homeless bum in San Francisco can wander into a local convenience store today and face a similar selection, so in that respect a bum in San Francisco is better off today than anyone in the Soviet Union 30 years ago.
[…] It didn’t matter if you were Gorbachev or Yeltsin, you literally couldn’t wander down the street and go to a store with 5 varieties of beans, 20 brands of cereal, etc. There wasn’t even a warehouse anywhere in the entire country that had that variety. If you as secretary general of the communist party had a sudden hankering for, say, bran flakes or frosted fruit loops, you couldn’t even command an underling to get it for you because if they hadn’t been made this month (or imported this month) there were no bran flakes or fruit loops. Oh sure you could shoot the deputy commissar in charge of bran flakes for not getting you any but that still didn’t get you any bran flakes.
Here in rural Japan I sometimes bitch about the unavailability of, say, certain sorts of cheese, or Marmite, or humus in the local shops. But thanks to people at Amazon, Rakuten and other online shops I can order any of them, even Marmite, and it shows up in a couple of weeks at worst and often the next day. And of course more normal things for this region (e.g. fresh fish, wagyu, rice, soba …) and general produce are available at multiple local supermarkets. This is a level of choice that rivals, in some ways beats even, places where the superrich live like Martha’s Vineyard or Atherton and it absolutely blows away the choices available for the richest people in, say, Kazakhstan let alone Sudan or Haiti.
But there are degrees. The plutocrat in Manila, say, can get pretty much everything I can get and likely more. The homeless bum in San Francisco can get something similar to almost everything I can get but since he likely lacks an amazon account more exotic produce is unavailable to him unless he makes his way to an upmarket grocery a dozen miles north or south. And so on.
I’ve been talking about food, because it’s easy to grasp, but the same applies to other things too. On the whole all sorts of other stuff from Ferraris to pencils via Apple iDevices and the like are equally more available and more affordable to all the people in rich countries even the relatively poor ones - I know people who have bought (well) used Ferraris with loans and a lot of fast talking in California for example.
So, on the whole we want to live in rich countries even if we are not rich ourselves because on the whole richer countries are also more pleasant. If you want a vivid illustration of that consider the border between Switzerland and Italy. The Swiss canton of Ticino looks very Italian but it is also clean, lacking in graffiti and generally safe. By contrast, just a few miles away (even yards away in the case of Chiasso) actual Northern Italy looks pretty dirty and unsafe and as you (if you take the train) wend your way southwards via Milan and Rome towards Naples it gets steadily worse.
But it is worth noting that a significant fraction of what makes richer countries more pleasant isn’t purely money. If it were then Qatar would be a paradise (spoiler note: it’s not even close). Things that make places pleasant and rich are cultural things like trust, equality, the rule of law, property rights, some (but not excessive) government regulation on safety and the environment and so on. Rich but unpleasant counties (hi Qatar) have poor records on trust, safety, property rights and so on. On the whole the correlation between pleasant countries to live in and rich countries is high but as with personal wealth vs national wealth it isn’t 100%.
But personal wealth as perceived by individuals is only loosely correlated to national GDP or GDP per capita.
Financial Assets Mostly Don’t Register
For most people, most of the time, the capital value of things like their house and their brokerage account don’t matter. What matters is the income (from the brokerage) and expense (house mortgage, taxes etc.). If a person can’t afford the $2000/month mortgage payment it doesn’t matter whether its a 1% rate loan on $1 million or 10% on $100,0001. If a pensioner gets $2000/month income from investments it doesn’t matter much whether that’s a return on a $5 million investment or a $500,000 one as long as it’s enough to pay for what is wanted.
Hence when people like Noah Smith go on about the stock market tanking most people don’t care.
At some point, if stock market crashes lead to bank runs and other financial disasters they will, but whether some index is at 35000 or 45000 really doesn’t register.
In fact if you don’t need to liquidate them for the money right now, the value of almost every asset is irrelevant to most people. What people care about is income, expenses and critically the purchasing power of disposable income.
As a personal example, my gross salary (in fact probably my net salary post taxes, health insurance etc.) is lower here in ruralish Japan than it would be in Southern California. My gross salary is probably close to half what I could get in San Diego county if I hustled for a job, given the current JPY:USD exchange rate. Even a decade or so ago when the exchange rate was less bad, the salary I could command in San Diego would be significantly more than my current salary here. But, and the wife and I did the sums then, our net income once we paid for basic things like housing, taxes, health, property etc. insurance, utilities and so on was greater 10 years ago here rather than there. As the US has suffered far worse inflation since then, despite the decline in the yen I’m fairly sure that I have more money to spend on clothing, food, and leisure activities here that I would there and the cost of said clothing, food, and leisure activities is lower here so I get far more bang for my yen than I would for my buck. In other words, in all the ways that matter to the average person, I am better off here than my imaginary counterpart who stayed in San Diego.
In similar fashion, along with any number of workers in Silicon Valley in the late 1990s I was technically a millionaire based on my stock options and the like. For any number of reasons, that include in my case a strong desire not to pay taxes, I and many others failed to turn those paper millions into actual cash and they disappeared in the .com crash of 2000/2001. For me and many others that loss made precisely no difference to our quality of life. What did make a difference for some was employers going titsup.com and then again for some, problems selling houses or making mortgage payments that had been fine based on previous expected levels of income. But for most of us, even those like me whose employers did eventually fail, the ability to keep one’s job or get a new job at a similar salary meant no impact on disposable income.
Finally, regarding the stock market. People with 401ks, IRAs and other savings and pension schemes may care about the health of those savings and hence the stock market. But many people in the US and around the world, particularly younger people and those in blue collar jobs, don’t have any exposure to the stock market. If the stock price of every company in the Dow Jones went to zero it would make no difference to them because they don’t even have indirect ownership in any stocks at all. Moreover, many have extensive debts - student loans, car leases, credit cards - and paying off those debts now is far, far more important than the pension they might get in a few decades time.
The Price of Fuel and Eggs
This is why the price of eggs and fuel matters. For most people, even relatively well off ones, the price of daily necessities like eggs, fuel and so on makes a big difference to their disposable income. For people on the margin it makes an enormous difference because it can mean the difference between paying off the credit card each month or failing to and having a debt at 22% APR rack up instead.
So when interest rates rise and inflation happens, since wages always lag inflation people get poorer. It doesn’t matter if GDP or even GDP per capita also rises. People are poorer until prices come down or post-tax wages rise. Relatedly, if GDP rises but GDP per capita doesn’t because of an increase in residents that doesn’t help.
This isn’t just a US issue. The headlines sound silly but while the US has been worrying about the price of eggs, on this side of the Pacific the worrying is over the price of cabbage2 and (more critically) rice.
Rich Countries vs Poor Countries Redux
Putting it all together a better way to try and measure the actual wealth of a nation is something like this.
For the residents, richer countries provide them with more disposable post-tax etc. income than poor ones and more freedom to spend that money on the products or services of their choice (saving/investing it is one of the choices). Freedom not only encompasses choices of what to spend it on, but when to spend and having the free time to enjoy the results as desired.
So we need to measure two or three things.
The first is the breadth of market offerings, including the variety of products and services and some measure of availability by time of day/week etc.
As we do so, we immediately see what we know instinctively, that some GDP rich countries like Qatar start to look less rich because entire swathes of product and service offerings (e.g. alcohol, christian churches) are totally unavailable. It need not be national, as regions in a country will differ in wealth. Opening hours makes places where shops shut on Sundays (and sometimes even other days) and are not open as much of the day score poorly and the actual existence of local shops is also relevant. This can of course be regional, or even hyperlocal - the “food desert” problem is not universal in the US, frequently you just need to travel a few miles and there are shops ready to serve. Rural communities tend to look poor on this because you often have to drive miles to buy stuff and the closest shops tend to be small with limited selection. But this, again should not surprise, there’s a reason why humans seem to prefer cities and suburbs.
Next there’s the net free income. That’s income after taxes, mandatory insurance, mortgage/rent, utilities and some base amount of food. It should include costs of any commute to work - season tickets, vehicle running costs. Essentially what I’m trying to measure here is the margin of income that a person has that they can choose to spend on a bigger car, better food, restaurants, movies, sports games and so on. It’s the pot of household money that shrinks when food and fuel prices rise and so on, and it needs an adjustment for purchasing power - based perhaps on something simple like the big mac index.
Finally there’s time. This is the amount of time not spent working or commuting to work.
Calculating some of these statistics may be hard, deciding the weights for combining them is going to take thought if there’s to be a single statistic instead of a triple but the end result is going to give a lot better idea of actual prosperity as experienced by citizens/residents that GDP
Numbers for example only. I have not calculated the actual rates
