These income-based class analyses don't make a lot of sense for reasons we've discussed in many past threads. Let's suppose that for the last 10 years you've earned $500,000 per year and saved most of it. Then you retire and your income drops to zero. Are you now low class? Or consider the reverse, where you've earned $20,000 per year your entire life, but then happen to earn $300,000 one year, perhaps from a personal injury settlement. Are you now upper class?
I think a more useful classification scheme for socioeconomic research would be based on asset level, but not necessarily the person's
actual asset level. Instead, it might consider the asset level that the person would have if they had lived frugally for their entire live, which I'll term "imputed asset level". In practice, we would use a proxy for this value, probably based on the person's age, lifelong income history, education, and other factors, all questions that could be asked on surveys and plugged into a heuristic formula to approximate the person's imputed asset level. I have never seen any research that used this approach to date.
Once some data was collected on the distribution of approximate imputed asset levels in the public at large, I speculate that the class bands would fall roughly into these four broad groups:
- Lower class: Imputed asset level is close to zero. Person is totally dependent on a job or other source of income (such as government programs).
- Middle class: Sufficiently high imputed asset level to live without income for maybe 5-10 years, but not forever.
- Upper middle class: Sufficient imputed asset level that the person never has to work again, but only if they maintain a modest lifestyle.
- Upper class: Wildly extravagant imputed asset level, such that the person can do pretty much anything without money as a barrier, such as constructing space shuttles, building hospitals, casually donating $20 million to a university here and there to create scholarships, etc.
The exact thresholds would have to be obtained from data, but overall this system would be far more useful than the current approach of pretending that current income level solely defines "class". If the current income is sufficiently high, like say $500,000 per year, that pretty much guarantees an imputed asset level above "lower class", so current income level is certainly related to class, but it doesn't define it alone.