How Rich are Americans, by Age Group?

How Rich are Americans, by Age Group?

In September 2017, the U.S. Federal Reserve issued a report based on the Survey of Consumer Finances. This survey was based on 6254 families, and the Federal Reserve report broke down the data by age group.

Note that this survey includes a house as part of a household's assets; page 14 of the report specifies "Declines in house prices in particular had a disproportionate effect on families in the middle of the net worth distribution, whose wealth portfolio is dominated by housing."

I've created a graph to illustrate this information, using LibreOffice Calc, a free and open alternative to Microsoft Excel. Here's the graph:

Here's a link to the Federal Reserve report

Now this is a great opportunity for a lesson in statistics and finance. You may be wondering why the mean (represented by the blue bar) is so much higher than the median (represented by the orange bar).

Example Data Set I: 1, 2, 3, 5, and 14

You've probably learned how the mean works: add all the numbers, and divide by the number of data points. For example: 1 + 2 + 3 + 5 + 14 = 25. Since we're adding 5 numbers together, divide 25 by 5, and there's your mean: 5.

But the process is a bit different to find the median. To get the median of those same numbers: first, make sure they're lined up in ascending order. So if your data set goes 5 + 3 + 14 + 1 + 2, make sure to re-arrange the list so it goes from smallest to biggest. Once the numbers are properly arranged (1 + 2 + 3 + 5 + 14), you strike out the first and last numbers, and you keep doing that until you've arrived at the middle number. In this case, drop the 1 and the 14, then drop the 2 and the 5. You're left with 3 – that's your median.

But what if there are two numbers left in the middle? You can't cross them both out! In that case, take the mean of the two numbers that are left. If we add another 14 to our list above, then you'd be left with the 3 and the 5. To get the median, add the 3 and the 5 and divide by two: the new median would be 4.

So why the math lesson?

To answer that, we'll go back to the initial question: why is there such a big difference between the blue bar and the orange bar?

Let's return to our math lesson. What was the mean?

The mean was 5.

What was the median?

The median was 3.

Why are these numbers different? After all, they're coming from the same data set!

Because of how they're calculated. The mean is more heavily influenced by extreme values, called "outliers." 

To illustrate this, go back to our example data set and replace the 14 with a 9, to generate data set 2:

Example Data Set II: 1, 2, 3, 5, and 9

Example data set 2: 1 + 2 + 3 + 5 + 9 = 20. Divide by 5. The new mean is 4.

But what's the new median? 1, 2, 3, 5, and 9. Drop the 1 and the 9, then drop the 2 and the 5. The new median is 3, which is the same as before.

Okay, refer back to the original data set. Now, replace the 14 with 89 to generate data set 3:

Example data set III: 1, 2, 3, 5, and 89

Now what are the mean and median?

Mean: 1 + 2 + 3 + 5 + 89 = 100. Divide by 5. The new mean is 20.

Guess what the new median is? Drop the 1 and the 89, then drop the 2 and the 5. What are you left with?

Yep. The median is still 3.

This example demonstrates why the median is the preferred method of reporting financial data. If you do a survey of people's annual salaries, and you've got 99 typical people, and then the 100th respondent is LeBron James and his $38.3 million annual salary, guess what's going to happen to the mean salary in your survey?

Right – LeBron's salary is going to distort the picture of how much people get paid. I've never met a welder or HVAC contractor (or even a lawyer or physician!) who makes $38 million a year!

But when you're conducting a survey of over 6000 households – like the Survey of Consumer Finances – you just might get some CEOs and investment bankers with 7-figure annual salaries, and likely at least 6 figures sunk into savings and investments.

Sure enough, refer back to that graph and take another look at the difference between the mean (which is heavily affected by outliers) and the median (which is affected only a little by outliers). Any guesses what's happening here?

Just like Example Data Set III, you can bet there are a few families in this survey with a very high amount of wealth. Judging by the massive difference between the means and medians, there are probably a couple families in here with assets over $10 million! Hence, the massive gaps between the blue bars and the corresponding orange bars.

*Side note: I've never been a fan of the term "net worth," because it implies that your worth depends on how much money you have, as though poor people are inherently worth less than rich people.

The term "net worth," of course, is in standard usage and shouldn't be taken to indicate someone's worth as a human being. But still, I prefer the term "net assets" because "assets" doesn't have the same connotations as "worth." I will therefore use the term "net assets" throughout this post.

Take a look at the "under 35" group. There's already a sizable gap between the mean and the median; most people in this age group have pretty modest assets, as indicated by the median.

The median is defined such that half of the under-35 people in this survey have less than $11,000, and half of them have more than that. Judging by the gap between the median ($11,000) and the mean ($76,200), this age group already has a few who are far wealthier than the typical person.

Despite my own 8 years of higher education and extremely low earnings during that time, I'm still far above the median for this age group. During my first year as a professor, I managed to save nearly 50% of my pre-tax salary.

If you've read The Millionaire Next Door, you won't be surprised by this. As authors Stanley and Danko observed, many educators are "prodigious accumulators of wealth," due in no small part to their frugality. So in that regard, I guess I fit right in with many members of that occupational group.

After all, they don't call me the "Froogal Stoodent" for nothin' 😉

But that sort of savings rate is not typical for many people. How many people do YOU know who are struggling to make ends meet? Who are worried what will happen next month? Or, since the current COVID-19 environment has shut down huge swaths of the economy, how many people will have their homes foreclosed or their cars repossessed because their incomes suddenly stopped with little warning?

At times like that, people tend to look to the government to ease the financial burden. Perhaps they get angry about their situation, railing about how the rich get richer while the poor get poorer.

This survey is meant to capture the net assets (or "net worth") of a variety of Americans, including everyday folks with unremarkable salaries. So according to the data in this survey, do the rich get richer as the poor get poorer?

To answer this, let's look at the gap between median and mean for each age group, expressed as a percentage. We'll take the median and divide by the mean, which gives us a ratio. We can express this ratio as a percentage by multiplying by 100:

Under 35: 11,000/76,200 = 0.1443, or 14.43%. This means that the median is 14.43% as high as the mean.

Age 35-44: 59,800/288,700 = 0.2071, or 20.71%. This means that the median is 20.71% as high as the mean. The gap between the median and the mean is actually smaller than for the "under 35" group.

Age 45-54: 124,500/727,500 = 0.1711, or 17.11%. The median-to-mean gap has gotten bigger.

Age 55-65: 187,300/1,167,500 = 0.1604, or 16.04%. The gap has gotten bigger.

Age 65-74: 224,100/1,066,000 = 0.2102, or 21.02%. The gap has gotten smaller.

Over 74: 264,800/1,067,000 = 0.2482, or 24.82%. The gap has gotten smaller.

As you look from youngest to oldest age groups, the median-to-mean ratio goes from 14% to 21% to 17% to 16% to 21% to 25%. Recall that the mean is more strongly affected by outliers than the median. So the wealth gap appears to increase throughout the working years, starting with the 35-44 group, but then decreases upon retirement.

So what does that mean?

Well, it's not clear. Data can tell you what but not why. This information tells us what's happening with people's money, but not why things are happening that way. There are multiple possible explanations for this pattern.

One possibility is that those who have more money are likely to handle it better. Remember: we have numerous cautionary tales from celebrities like M.C. Hammer, who earned more than $30 million in one year in the 1990s! Most people don't earn that much in an entire lifetime, let alone a single year!

So with over $30 million in a single year, as well as millions more in other years, Hammer must have been set for life, right?

Wrong. He endured a highly publicized bankruptcy. This goes to show that if you can't handle your money, it doesn't matter how much you earn. Not even if you're in the top 1% of the top 1%, like M.C. Hammer was!

Naturally, as long as you can prevent lifestyle inflation, a higher salary will help you save money faster. But your mindset and your priorities shape your net worth far more than your income level.

During retirement, the median continues to increase while the mean decreases, yielding a diminishing gap. This is an odd pattern: it would appear that the wealthiest people appear to live a more indulgent lifestyle in retirement, while the less-wealthy either continue to work – thereby generating income – or tighten the belt in retirement, surviving on social security and pension benefits.

It could also mean that the wealthier people may retire earlier, thereby drawing on their retirement money earlier and diminishing their nest eggs.

Another possibility is that this reflects a general move among the wealthier set from high-yield investments like stocks to lower-yield investments like bonds, though by itself, this is not a likely reason for net assets to diminish unless that's also paired with early retirement.

So as we've seen, the data tell us what but not why. To get definite answers, you'd have to dive deeper and include some additional questions to ascertain people's spending habits, earning habits (second job, overtime, etc.)

What you can take from this analysis is that the rich tend to get richer until retirement. And by that point, their tendency to save has paid off and allowed them to draw on their retirement funds to allow a comfortable retirement, if they so choose.

So how do you compare? Are you on target with the median for your age group? How about the mean?

***
Support the Froogal Stoodent blog by checking out these laptops or these headphones;
anything you buy will generate a small commission—at no cost to you! Thanks!