Is median household income a poor measure of actual affluence?

[they don't love you like i love you]

Yes I'm not fond of it. It's very dated and might've been a good or relevant measure in the 50s, but it isn't today. The problem is that it artificially deflates the median incomes of places that have lots of singles. For example Minneapolis is majority single and plurality never married. It has a median income of $45k. This is actually below the national numbers, but for such a city pretty high. And I'd probably qualify as lower middle class where I'm at now, but if I married someone at my job we'd be more affluent than the median income in the country.

Per capita income is a better measure in my view.

[muon2]

The presumption is that households can pool expenses. Since housing costs are typically the largest single expenditure, and food prepared at home also benefits from shared resources, household income is a better measure of what would be available for spending on non-essential goods and services. Per capita income also distorts spending power in areas with a lot of children since they do not usually consume as much as the adults.

The example you give is a good one to use. At $45K you might be expected to put $15K into housing expenses. That would leave $30K for other expenses. Two people with $90K don't need to spend twice as much on housing, but more like 1 1/2 times to get the same living space per person. That might mean spending $22K on housing, leaving $68K on other expenses or $34K per person. That makes the household substantially more wealthy than the two individuals living separately.

[opebo]

Yes, I agree that per-capita is far better.  The example muon gives above mentions housing - however in fact most couples must (as a part of the progression towards child-bearing) acquire much larger and more elaborate housing than the two single apartments they leave behind.  Not to mention that this child rearing adds another enormous layer of expenses of all sorts.  On the whole I would say that two singles making $45k each actually become poorer by marrying, buying a house, and rearing children on their pooled $90k.

[jimrtex]

The poll question is poorly worded.  It is expecting a "yes" response for something that would be considered negative ("poor measure").   There is juxtaposition of "poor measure" and "affluence" when "poor" is the usual adjective used for someone who is lacking affluence, and the added word of "actual" in front of "affluence".

Don't you agree that it would be much better to reword the question in a way that would avoid confusing or leading respondents?

[DC Al Fine]

Yes I'm not fond of it. It's very dated and might've been a good or relevant measure in the 50s, but it isn't today. The problem is that it artificially deflates the median incomes of places that have lots of singles. For example Minneapolis is majority single and plurality never married. It has a median income of $45k. This is actually below the national numbers, but for such a city pretty high. And I'd probably qualify as lower middle class where I'm at now, but if I married someone at my job we'd be more affluent than the median income in the country.

Per capita income is a better measure in my view.

The biggest issue with household income is that people assume that the average household stays the same over time, which it doesn't. More single people would by definition lower household income ceteris parebus.

[All Along The Watchtower]

Looking at an area's household income distribution might be helpful too, no?

[Benj]

Per capita income has its own problems in being skewed by a few extremely high earners. So, while somewhere like Manhattan is definitely underestimated in median household income due to the high concentration of singles, it's also definitely overestimated in per capita income due to the high concentration of multi-millionaires and billionaires.

[opebo]

Per capita income has its own problems in being skewed by a few extremely high earners.

Median salary or median hourly wage would be the way to go - something like that.  Though even median isn't nearly as important from a social policy standpoint as the 'income of the bottom quintile' would be.

[Stranger in a strange land]

yes, it is, and honestly even per capita income is a poor predictor of affluence. For example, a single guy making 50K per year where I live (a pretty expensive area) is doing very well, but his co-worker who's also making 50K but is the sole provider for a family of four is probably just barely getting by. However, in a rural area, a family making 50K would be a lot more comfortable and be able to save and invest a lot more.

A better predictor would be what a household can buy in the community where they actually live.

[muon2]

yes, it is, and honestly even per capita income is a poor predictor of affluence. For example, a single guy making 50K per year where I live (a pretty expensive area) is doing very well, but his co-worker who's also making 50K but is the sole provider for a family of four is probably just barely getting by. However, in a rural area, a family making 50K would be a lot more comfortable and be able to save and invest a lot more.

A better predictor would be what a household can buy in the community where they actually live.

Disposable income which subtracts all taxes from income is easy to find online and gets in the direction you suggest. Discretionary income then subtracts all compelled payments (bills) which would be a better measure of the type you suggest. However, that number is not specifically tracked by the Bureau of Economic Analysis at the Dept of Commerce. It is sometimes published by other groups using a variety of data sources, but is often proprietary, since that is the kind of data that businesses use to make decisions such as where to locate new stores.

[Torie]

Age has something to do with it as well, to wit how much human capital you have left (young professionals have a huge amount), and net worth (olds have more).  If you have a lot of human capital, you can borrow more, and spend more (knowing the bigger bucks will be rolling in soon), and in general have a more affluent lifestyle. And youngs have far less in health care costs, and insurance premiums (now however destined to balloon up - unfairly in my view - but I digress). And then there are college educational costs, now horrifically high. No one measure captures it all, obviously.

[opebo]

Regardless of the measure of actual affluence, as a political decision it makes sense to set the wages of each individual job at an ample sufficiency to cover the costs of raising a 'typical' family - say one worker for two spouses and 1-3 children.  To do anything else is to invite social problems.

[memphis]

How does the census measure household income in the case of people who are living together but not in a relationship? It is very common for two or more single 20 or 30 somethings to split an apartment for the sake of finances. The IRS would certainly count them as separate households, but the census just sends one form to each address.

[Benj]

How does the census measure household income in the case of people who are living together but not in a relationship? It is very common for two or more single 20 or 30 somethings to split an apartment for the sake of finances. The IRS would certainly count them as separate households, but the census just sends one form to each address.

One household for MHI. "Household" is determined solely by housing unit. You're one household if you live in a single housing unit together, even if you're not in any kind of romantic relationship (let alone married). You're two households if you live in separate housing units even if you're married and file taxes jointly.

"Household" isn't actually a term used by the IRS, of course.

Although one might at first think this is problematic, it actually helps offset the other problem with MHI, namely that it tends to underestimate places with lots of singles living alone. Since a fair number of singles will often live with roommates, they help offset the lack of married couples.

[Smid]

The Australian Bureau of Statistics uses SEIFA - Socio-Economic Indexes for Areas. It ranks areas according according to advantage and disadvantage. Obviously income is one factor used in the calculation, but it factors in other variables. More information, including a Technical Paper which sets out the variables and reasoning, is located here.

The ABS defined advantage as:

You must be logged in to read this quote.

From memory, it includes things such as home access to broadband as well as more obvious measures, such as income.