Look at this chart (Figure 6):
Anything look odd?
You may be saying, “Josh, what the heck is this? This is a waste of my time!” But let me tell you my doubting friend, it’s charts like this that researchers use to determine American’s ability to retire. Let’s go into exactly what we see here.
The SCF is:
“Wealth data from the SCF are widely regarded as the most comprehensive data available for the United States.” Part of the data set of the SCF is a self-reported survey of Americans which “allow the respondents to reference their tax forms when answering the income questions.”
All of the above information, by the way, comes from an IRS research paper which you can find here.
What we’re doing here is comparing IRS data, from actual tax returns to SCF data which is considered the most comprehensive data in the U.S. If you look at the Figure 6 again, you will see large discrepancies between these two surveys.
Notice the huge difference between Taxable Interest, Dividends, Capital Gains and, very interestingly “Pensions, annuities, SS” (which is Social Security). The IRS data picks up 60.7% MORE income in that category than does the SCF!
In the other categories the IRS data picks up well over one-third more income than does the self-reporting of the SCF. Even more interestingly is the “other income” category. The SCF reports nearly 50% more income than does the IRS pick up.
Finally, though, one may be inclined to say “Josh, I get it. This is odd. But look at the TOTAL row. Notice that’s only a 6.1% difference between the two. So, not really a big deal.”
Initially, I thought the same. But then I noticed that income discrepancies are HUGE when it comes to the types of income most retirees have; Interest income, Dividends/capital gains and certainly Social Security/annuities/pension income.
Most retirees don’t have business income or wages and salaries. So, discrepancies there are not that important regarding retirement planning. But the other areas most certainly are.
And what Figure 6 shows is that the self-reporting data is WAY off from the actual IRS data. We are talking BILLIONS of dollars the IRS reports as income from American taxpayers who do not report that on the SCF surveys.
You can, and actually SHOULD, read the article to find out why this is the case. It’s nothing nefarious, it’s just simply in the way the SCF asks the questions. Which is something I’ve been harping on for years. See video here.
But if you wanted to create a retirement crisis what data set would you use to “prove” that American retirees don’t have enough money? Would you use the ACTUAL IRS data which has Americans with BILLIONS more income, of which they definitely are taxed on? Or would you use the SCF data?
You can even couch your concern by saying something along the lines of: “In using actual American survey data we see they don’t have enough income. After all, who better knows their own income than actual people who report it?” Yet, that self-reporting is WRONG! We know this factually because it’s the IRS data that is used to determine taxes due. And no one is going to OVER report income to the IRS if they actually don’t receive that income. No one does that.
So, if you want to scare people, simply show them the SCF data and ignore the IRS data. And that’s how you create a retirement crisis.