If you are reading this, you probably have the impression the US is among the richest countries in the world (per person), but not necessarily at the top and certainly not substantially ahead of everyone else. At least that was my impression from everything I’ve read and heard. If you look at GDP per capita statistics, which are usually what is cited in the media, this would seem to confirm that impression:
Source: Wikipedia [1]
Sure, the US is close to #1, but the average GDP of the Scandinavian countries is not too far off and Luxembourg, Switzerland, Ireland, and Norway are all ahead. But what if I told you the US actually DOES have a substantially higher (material) standard of living than the rest of the world? When we look at a different metric called adjusted disposable household income (ADHI):
Source: Wikipedia [2]
Wow! Now the US isn’t just #1, it’s almost 20% above the 2nd highest country. When we compare the standard of living in other countries, which one of these metrics should we use? I’ll answer that question in a bit. But first, does it even matter?
To some extent, single metric cross-country comparisons aren’t particularly useful if they aren’t part of a more rigorous statistical analysis. For example, deciding whether or not Norway has a higher standard of living than the US doesn’t increase our understanding of how the world works in a meaningful way on its own. Additionally, GDP and ADHI are very imperfect proxies for a country’s standard of living since both fail to capture the distribution of income as well as differences in health, happiness, leisure time, unpaid work, and environmental stuff.
That said, single-metric cross-country comparisons ARE frequently made and I think there are some interesting implications of GDP capturing the narrative over ADHI.
What Looks Different With ADHI?
1) Anyone who has read about health care prices in the US has probably seen graphs like these:
Source: Peter Hilsenrath [3]
Source: RCA [4]
The US looks like a massive outlier in how much money we spend on healthcare, and yet our life expectancies are actually worse than most countries in our income neighborhood. A graph like this is usually shown as evidence that the health care system in the US is dysfunctional. But if ADHI is a “better” metric, maybe the US doesn’t actually have uniquely expensive healthcare (which isn’t to say there aren’t many issues with the system):
Source: RCA [8]
Source: RCA [7]
Here the US is right in line with what you would expect based on its Actual Individual Consumption (AIC), assuming a non-linear relationship between income and healthcare spending (which the GDP graphs don’t do). AIC is very similar to ADHI (correlation=0.98), so don’t worry about the change in acronym. If you’re interested in healthcare spending, this post was inspired by several really interesting (and VERY long) articles by Random Critical Analysis that question whether the US is really a health spending outlier (see https://randomcriticalanalysis.com/2016/09/25/high-us-health-care-spending-is-quite-well-explained-by-its-high-material-standard-of-living/, https://randomcriticalanalysis.com/2018/11/19/why-everything-you-know-about-healthcare-is-wrong-in-one-million-charts-a-response-to-noah-smith/ ).
2) It’s a common talking point that Scandinavian countries are just as successful as the US despite higher tax rates, more regulations, and fewer hours worked. Obviously average material standard of living may not be the most important metric for comparing how people fare in different countries, but the ADHI numbers contradict the Scandinavia argument that you can have more "government involvement" without any tradeoffs in economic growth/success. As I mentioned above, we probably shouldn’t be using GDP or ADHI to make political statements since there are so many differences between countries (ex. it’s possible that higher social capital in Scandinavia is the key causal variable in their relative success). But if we are going to make these comparisons, we should have a good understanding of the “true” situation.
3) There has long been a debate about if “American Exceptionalism” is a valid idea. I don’t really care about this, but surely knowing whether the US has a far higher standard of living than the rest of the world is relevant here.
Ok, my intent in thinking of implications wasn’t to show that people on one side of the political spectrum might be more interested in this than others! I’m sure there are some other implications going the other way (ex. If we’re doing so well overall, maybe that suggests we can afford to flatten the distribution a bit and spend more on public services and infrastructure). Regardless of implications, I think the disparities between GDP and ADHI are very interesting on their own, especially given how GDP is THE METRIC used when discussing economic issues and the US’s place in the world and few people have even heard of ADHI or AIC.
Alright sure, so what goes into GDP and ADHI?
The inputs to these metrics are actually very simple.
GDP is:
Household consumption: All the stuff people buy for themselves
PLUS
Investment: People buying houses or businesses buying equipment (but not people investing their money in the stock market)
PLUS
Government spending: Spending by government, duh
PLUS
Net Exports (exports-imports): This captures products produced in a country but consumed elsewhere
ADHI is even simpler:
Income: Earnings + income from capital
PLUS
Transfers: Think public education, SNAP and Social Security in the US
MINUS
Taxes:
And why does the US look so much better using ADHI relative to its GDP?
Across the OECD, the US has the biggest positive gap between GDP and ADHI. More importantly, the four countries with GDP higher than the US all have a large negative gap between their GDP and ADHI.
Source: RCA [9]
There are several ways GDP and ADHI can diverge:
Foreign-owned company profits and foreign worker incomes: Included in GDP but in ADHI they would be assigned to the country where the income/profit eventually goes
Ireland and Switzerland have a lot of primarily foreign-owned firms and workers because of their weird tax and finance rules
I think the US bump is because there are a lot of US firms outside the US
I’m less certain of this but I think Luxembourg has a lot of foreign workers (and sorry but who cares it’s Luxembourg)
Ireland’s ADHI GDP divergence is fairly recent:
Source: BSL [6]
Remittances and Foreign Aid: Included in ADHI
These are fairly small for developed countries but a bigger deal in the developing world
Government Savings: Included in GDP
This explains Norway’s divergence. Having a massive sovereign wealth fund (worth ~$200,000 per citizen) is a good thing and should generally result in higher long-term consumption, but the impending oil decline (and oil's importance [20% of GDP, love a good double parentheses, good luck following this] to the Norwegian economy) means they will likely need use their savings to maintain their current standard of living
Corporate Savings (aka retained earnings): Included in GDP
This is somewhat different in the countries above but not enough to explain the divergence we care about here
“Collective” Government Spending: Included in GDP
This would be all government spending that can’t be directly tied to people (ex. Infrastructure, prisons, national defense)
Somewhat surprisingly (to me at least), The US spends a higher than average percent of GDP on this stuff (see below, presumably US would be about on trend for ADHI instead of GDP), so this doesn’t explain why the US stands out on ADHI
Source: RCA [9]
Which one makes more sense when we talk about a country’s standard of living?
So we’ve shown that the US has a higher ADHI, primarily due to US assets that are located outside the country, and other high-GDP countries have various financial quirks that result in lower ADHIs. But which one reflects material well-being better? ADHI should be preferred for two reasons:
A large panel of distinguished economists recommended using ADHI over GDP when we are interested in standard of living:
"Recommendation 1: When evaluating material well-being, look at income and consumption rather than production…
Material living standards are more closely associated with measures of net national income, real household income and consumption – production can expand while income decreases or vice versa when account is taken of depreciation, income flows into and out of a country, and differences between the prices of output and the prices of consumer products." [6]
ADHI is a much better predictor of almost everything we associate with standard of living (everything above the line is a better fit with AIC):
Source: RCA [9]
Source: RCA [9]
Lots of important stuff in these graphs, including the Social Progress Index, life expectancy, democratic quality, all are better predicted when we use AIC instead of GDP.
Sure, so you accept ADHI as a better metric. You might still be skeptical that the US is really that much wealthier than every other country. Let’s look at some consumption data:
Source: RCA [9]
Unfortunately this is from 1999, but you can see the US had far more domestic appliances than other rich countries. Computers, VCRs, microwaves, cars, and dishwashers seem especially notable. It seems unlikely these all reflect Europeans’ differing tastes and much more likely it reflects Americans having higher incomes.
Source: RCA [9]
Ok there’s a lot going on in this one. If the US is really so well off, we’d expect to see them outspending other countries in most of these areas (especially where the slope of the line is higher). Sure enough, the US is close to the top in consumer services, consumer goods, restaurants/hotels, recreation/culture, transport, and home furnishings. Maybe this isn’t convincing on its own, but still more evidence pointing in the same direction.
If ADHI is so good, why isn’t it used more?
Wow, great question! I wish I had a good answer but I really don’t. I think its usage should be broken down between academia/research and the media.
Consumption (ADHI or AIC) is used to some extent by researchers, but I have no idea how often. Here’s a recent example:
Source: Twitter
It’s also worth noting that there are cases where it makes more sense for researchers to use GDP (I’m not sure when, presumably looking at more production type data), so we wouldn’t necessarily expect to see all researchers using ADHI and AIC. GDP is also available for every country (though it’s certainly not accurate for every country!), so it makes sense to use it in analyses where ADHI data is lacking (it’s harder to calculate and only available for ~35 countries). I’m fairly confident there are many papers that use GDP when AIC or ADHI would make more sense, but I can’t think of any easy way of determining how common this is. Frustrating, but alas that’s all I can say there.
What I am confident about is that ADHI and AIC are very rarely used in the media. I don’t think it’s to push a narrative that the US isn’t exceptional or anything like that, but it is still interesting that economists have started to move towards ADHI in the past decade without journalists following along. I’d be very interested to see a survey of media people and how they think the US material standard of living compares to other countries. I don’t think they’d do well!
Hopefully this post will (or won’t!) give you Gell-Mann Amnesia (https://www.goodreads.com/quotes/65213-briefly-stated-the-gell-mann-amnesia-effect-is-as-follows-you) the next time you read an article where GDP is mentioned to measure standard of living.
References
1: https://en.wikipedia.org/wiki/Disposable_household_and_per_capita_income
2: https://en.wikipedia.org/wiki/List_of_countries_by_GDP_(nominal)_per_capita
5: http://www.oecd.org/sdd/prices-ppp/OECD-PPPs-2011-benchmark-Dec-2013.pdf
No comments:
Post a Comment