This follows up on a recent post about the "lost" middle class in the US. Its interesting to compare the "middle" in the US to the middle in some of the rest of the developed world. That is what the article linked here does. Here is one excerpt:
"The green bar is the US income at poverty levels. So, this tells us that a person at 60% of median income in the US still has a larger income than the median household in Chile, Czech Rep., Greece, Hungary, Portugal, and several others. And the poverty income in the US is very close to matching the median income in Italy, Japan, Spain, and the UK."
It was somewhat surprising to me that being a "poor" household in the US is similar to being in the middle in places like Japan and the UK. The article suggests the US is still a pretty good place to be stuck in the middle even if its not what it used to be.
The article paints the picture from several different angles:
https://mises.org/blog/poor-us-are-richer-middle-class-much-europeFWIW
David
P.S. This article is being linked simply to share the data it presents. All politics was left at the door!
Comments
I spot checked a small sample of the article's data against the data in the Wikipedia, OECD, and Gallop links provided in the article. The Gallop data checked out against both sources. The OECD data checked out against the Wipipedia source. I didn't get the OECD source figured out enough to determine the median incomes there.....Anyway, I am assuming the author is accurately presenting the data in the article I linked above...
so if you have incomes of 11 people
1,1,1,1,1,7,7,7,7,7,7
median income is 7
and
6,6,6,6,6,6,6,6,6,10,10,10
median income is 6
what group would you want to be in?
https://www.mathsisfun.com/median.html
Yeah, I checked it cursorily against the Pew data, perhaps the best of the lot, which is why I deleted my comment of skepticism of anything involving the Mises Institute.
Not sure what the points are really, yay USA, exceptional us, go go go, no need to do anything?
I understand your point. The Gallop data is based on information from 2000 or more households in each country. I am not sure the minimum number for the OECD data but assume that each individual country's sample is "large". The distribution curve is presumably fatter towards the low end for most (all?) countries....how much fatter would vary by country. But, it seems to me focusing on the middle most household to compare these large data sets is a better way to represent each country's population than focusing on the average incomes would be. Median incomes probably provide a decent sense for how "most" people are doing in each country.
My personal sense is there's a lot that can be done to make the income distribution in this country more equitable. But that's where each person's individual perspective comes into play. We get another chance to point things in another direction next year.
My point in posting the article was just to offer a little global perspective regarding what it means to be part of the "middle class" in the US.
That is, that poverty for each country is 60% of its median income. That's a problem for a number of reasons; it's even a problem with the US data.
Here's a page from the Census Bureau entitled How the Census Bureau Measures Poverty. Nothing so simplistic as 60%. There's a link to an Excel spreadsheet giving 48 possible poverty thresholds, depending on family size and ages of family members. Eyeballing these figures (and not having the distribution of US household sizes in front of me), I would say that the average US poverty threshold shown in the OP graph is in the right ballpark.
This is only an average, highly dependent on the definition of household and distribution of households. It's also mixing a US definition of poverty with international (OECD) data. Why 60%? The UN uses 50% (see, e.g. this UN paper on child poverty, Figure 1B), as does the OECD.
If all this is too complicated, just look at Accipiter's sets. In the latter, no one is living in poverty, because no one is below 60% of median. Change four of the 6's to 4's, the median wouldn't change and still no one (including all the 4's) would be below the poverty line. Distributions matter.
U.S. versus Germany
http://www.numbeo.com/cost-of-living/compare_countries_result.jsp?country1=United+States&country2=Germany
Ann Arbor, MI versus San Francisco
http://www.numbeo.com/cost-of-living/compare_cities.jsp?country1=United+States&country2=United+States&city1=Ann+Arbor,+MI&city2=San+Francisco,+CA
And October 20 is World Statistics Day............just in time, apparently.
Regards,
Catch
But instead, the Germans have arranged their economy (somehow) so that major recurring expenses for foodstuffs (excepting beef) and rent is lower, despite their population density.
The inference I draw, is that while nominal US poverty income levels may seem well-situated vs other OECD median income levels, US industries seem rather more rapacious than their European counterparts at aggressively extracting that income, or said another way charging more to consumers vs. value provided.
Further 'catch22's' links does not include healthcare. Most (all ?) OECD countries except the US view healthcare as a necessity and do not permit the price-gouging and profit-maximizing behavior we do here. 'Medical bankruptcy' is a term which is a uniquely American phenomenon.
The point: 'income' cannot be looked at without considering expenses.
slate.com/blogs/moneybox/2015/01/05/america_s_poor_vs_the_rest_of_the_world.html
demos.org/blog/1/5/15/when-it-better-not-be-america
And this data excludes the healthcare benefits Europeans receive compared to the U.S.'s system. It also excludes the fact that Americans work longer hours and have less vacation time than Europeans for the income they receive.
From the Demos article: "Of course, disposable income is not a perfect indicator of which countries are using the best economic institutions. For one, different countries have different capacities to produce income, which is why I tend to prefer relative measures among developed countries. Additionally, disposable income fails to capture in-kind benefits, and therefore misses the wonderful free health care, among other things, available in many of the above countries. It also fails to capture differences in leisure time, which is much higher in many of the above countries. Finally (for now), it fails to capture the well-being boosts that comes from simply being in an economically secure environment as well as the increase in social cohesion and trust that is associated with lower inequality and lower social distance."
>> there's a lot that can be done to make the income distribution in this country more equitable.
I was attempting to be droll about American exceptionalism and the like.
similarly a lot of the middle class in europe does not have to worry about their kids college education.
WTF are we comparing people across continents/countries? can we please compare people within the united states? Or do we think we are done with that debate? the only reason to waste intellectual capital to look at other countries is to learn from the positives there. why focus on the negatives? to make us feel good about ourselves?
I thought the whole reason this discussion was started was to show how statistics can be used to produce spurious conclusions and ridiculous comparisons. Wasn't that the whole goal
>the only reason to waste intellectual capital to look at other countries is to learn from the positives there
from the comments we can see they do a better job with a few things : health care, education, quality leisure time, and price gouging in the medical field, etc. :;
"I thought the whole reason this discussion was started was to show how statistics can be used to produce spurious conclusions and ridiculous comparisons. Wasn't that the whole goal"
Amen!!
Did you know that the great majority of people have more than the average number of legs? It's obvious really; amongst the 57 million people in Britain there are probably 5,000 people who have got only one leg. Therefore the average number of legs is:
((5000 * 1) + (56,995,000 * 2)) / 57,000,000 = 1.9999123
Since most people have two legs...
---
Statisticians are like the drunk leaning against the lamp pole -- they are there for support not illumination.
---
What's the difference between a physicist, a mathematician, and a statistician?
The physicist calculates until he gets a correct result and concludes that he has proven a fact.
The mathematician calculates until he gets a wrong result and concludes that he has proven the contrary of a fact.
The statistician calculates until he gets a correct result about an obviously wrong proposition and concludes NOTHING, because the explanation is the task of the scientist who consulted the statistician.
Reply: I'm not clear of your comment was serious, or made with tongue-in-check?
I think its a common (but bizarre) tendency of Americans to NOT compare & contrast themselves, their society, policies, outcomes, circumstances with other nations. We can learn (well if we are capable of learning) from other societies' successes, as well as their failures. Ignorance of what other societies are doing/have done is NOT a virtue.
http://www.nytimes.com/2015/10/19/opinion/something-not-rotten-in-denmark.html
Some of your comments regarding the article I posted left me puzzled. I am hoping you can help me to better understand the points you were making. Here is what I am thinking:
1. Re your comment related to https://www.mathsisfun.com/median.html
These are "large" samples which have presumably been obtained using generally accepted randomized sampling methodologies (specifically stated to be 2000 plus data points for each Gallop sample; detailed data review process discussed at the OECD website related to their sampling techniques). So, it is my assumption there are numerous data points in close proximity to the median for each country, that the distribution curves are not "discontinuous" close to the medians, and that the medians presented do closely approximate the center point for the incomes of households within each country. Do you have reason to conclude this assumption is probably inaccurate?
2. Re: http://www.theatlantic.com/magazine/archive/2015/10/why-we-compete/403201/
My understanding of this article is that people focus on how they are doing relative to the peers in their "neighborhood". One extreme end of the spectrum would reasonably include an individual who is focused on their economic status only relative to the neighbors who live in their culdesac. The other extreme end of the spectrum would reasonably include an individual who is focused on their economic status relative to the entire "global neighborhood". The article suggests to me that most people live in fairly small "subjective neighborhoods". The Atlantic article does not appear to be addressing the validity of data presented in the article I posted. Do you agree? Rather, it appears to be a way of questioning the usefulness of making this sort of income based comparison between any populations where the people involved do not live in the same "subjective neighborhoods." Is this your basic point?
3. Re: "I thought the whole reason this discussion was started was to show how statistics can be used to produce spurious conclusions and ridiculous comparisons. Wasn't that the whole goal"
This comment reinforces my conclusion from #2 above that you may think it is useless to make comparisons involving people who do not live in the same "subjective neighborhoods". Is this correct? Are you also saying something specific about the methodology used to collect or present the data in the article I posted?
4. Re: "Did you know that the great majority of people have more than the average number of legs?"
This seems to me to support the use of the median rather than the average in this case! I don't understand the comment.
5. Your final comments seem to be variations of the saying that "Figures lie and liars figure". Is this a fair statement? Are these comments basically pointing towards the theme presented in the Atlantic magazine article or is another point being made?
6. One final question......Do you differ with my thinking that it can be useful to consider inter-country comparisons of this sort?
Thanks!
David
The little I know of statistics, you try to reject a null hypothesis and have a significance test. Descriptive statistics are purely descriptive statistics, the conclusions you draw from them are yours.
As far as richer, the only way I could personally tell was if I were poor in the U.S. and was adopted by a middle class family in Europe, then I could come back and tell you my anecdotal experience.
the points made by Lewis and a few others underlie the reasons why I cannot draw the same conclusions as you.
I'll have to leave it at that.
https://explorable.com/null-hypothesis
http://blog.minitab.com/blog/adventures-in-statistics/choosing-between-a-nonparametric-test-and-a-parametric-test
As far as the Atlantic article - I just posted it for you to read but not as a counter argument but as additional info.
Thanks for your response. I didn't draw any firm conclusions from the article. I just accepted that what the author stated may have some basis "in the data" and may provide some rough guidance for comparing the median incomes of the countries that were discussed. (I agree with you that personal experience would trump the kind of number crunching that produced the data used in this or any article. )
Here are a couple of comments:
1. The following quote from the NonParametric tests section of your minitab link discusses part of the reason it made sense to compare medians rather than means in this article:
"For example, the center of a skewed distribution, like income, can be better measured by the median where 50% are above the median and 50% are below. If you add a few billionaires to a sample, the mathematical mean increases greatly even though the income for the typical person doesn’t change."
2. The author of the article indicated the sample data was adjusted for Purchasing Power Parity. The article stated:
"When adjusted for purchasing power parity (PPP), the statistic allows us to make incomes comparable across countries that use different currencies and have different costs of living. This takes into account taxes, and social benefits paid to households."
The idea is that the OECD number crunchers adjusted the median income numbers between countries to account for some of the factors noted by Lewis including their differing health care systems. Its my understanding that PPP does not attempt to adjust for factors such as social cohesion and trust. (My personal philosophy is fairly egalitarian. I tend to agree with Lewis those non monetary factors are important to consider.) The author and the Wikipedia article state the Gallup data is also based on PPP. See the following for more information on PPP:
http://data.worldbank.org/indicator/PA.NUS.PRVT.PP?page=1
http://www.oecd.org/std/prices-ppp/purchasingpowerparities-frequentlyaskedquestionsfaqs.htm#FAQ7
FWIW,
David
Having traveled fairly extensively throughout Europe, and having asked a great many questions regarding comparative standards of living, I was less than impressed by the casual "taking into account taxes and social benefits". It's a complex situation, and I'd need to know a lot more about "taking into account" before giving much weight to that statement. There are plenty of differences between individual countries in Europe, never mind Europe and the US, so I'm also not impressed by lumping that whole area together. The same of course applies on this end- how do the living costs in California compare, for example, to South Carolina?
When we travel, it's to see how other folks are living. While others are out shopping for bling and assorted crap that can be found almost anywhere, we are visiting small grocery stores and larger department stores and merchandising centers to see what kinds of things are available and how much they would cost locally vs the US. And talking to people... that's the best part. My overall impression of the linked article was that it was pretty glib and shallow.
@Catch22 recently posted a great link which I found to be very interesting. It allows comparison on living costs between almost any two reasonably large cities in the world, and breaks down prices to the level of quarts of milk- SF: $5.23, Paris: $4.14, and loaves of bread- SF: $4.02, Paris: $1.45. The breakdown is very detailed- food, utilities, transportation, rent, etc. Two major areas that aren't covered are education and health care- and they are inextricably wound together with the European tax systems.
From personal knowledge the costs shown for San Francisco and various European cities are very close to the mark. Here is a preset link comparing San Francisco and Paris, for example. Give it a look- it's very detailed, informative, and accurate. Catch's original link compares Ann Arbor, Michigan (him) to San Francisco (me), and is also very interesting- Ann Arbor: a quart of milk, $3.48, a loaf of bread, $2.98.
Yes, the use of the median rather than the mean is justified in this sort of comparison. Unfortunately, with garbage in and garbage out for "data", use of the median isn't going to make any difference as far as accuracy.
I play golf with a number of Europeans here in SW Florida...one of my friends, from Slovenia, says clothing and gasoline (of course) is much cheaper here, but food is more expensive. She says golf is about the same, but once we played with a couple from Switzerland who said golf is very much an elite sport there...kinda like polo would be here.
Really difficult to compare "lifestyles" and "standards of living"...even between Americans....impossible between Americans and rest of world, IMO.
@little5bee- Yes, I absolutely agree with you on that. Interesting on your choice of McDonald's as a comparison index. The Economist does the same thing, and reports several times a year on their international "Big Mac" index. Actually not bad for a quick and dirty value comparison.
Check out Catch22's site that I mentioned above. Kind of fun, actually, and seems to be pretty accurate info, too. That's likely because they actively solicit input from people who actually live in the various locations (and actually buy bread and milk) rather than some BS set of data inputs "normalized" to make the results come out "right".
Explanation of chart from The Economist: For example, the average price of a Big Mac in America in July 2015 was $4.79; in China it was only $2.74 at market exchange rates. So the "raw" Big Mac index says that the yuan was undervalued by 43% at that time.