I remember talking to Walter Oi about pay in the early 1980s. He had an idea to write up a paper titled “What is Pay?” I don’t think he ever did. But here was the issue he led with.
The Air Traffic Controllers union had gone on strike and President Ronald Reagan, in his first year in office, had fired them while giving them no chance of being re-employed. Walter pointed out that some of them were within just a few years of getting pensions whose present value (the value of the stream of pension income they would get discounted back at a reasonable interest rate to the present–then 1981) was close to a quarter of a million dollars. In today’s dollars, that would be about $650,000. That’s what most readers–and writers–of this blog would consider a lot of wealth. “For them to go on strike, they must have put the probability of Reagan firing them at about one or two percent,” said Walter with his great chuckle and his sparkling smile.
His general point was this: we can’t simply look at someone’s current pay to really know what his pay is, at least in the sense that we want to know pay. If the person will get that pay for a long time, if the probability of getting fired is low to zero, and if the person qualifies at a relatively early age for an inflation adjusted defined-benefit pension, then pay is much higher than it looks.
This reminds me of a case I’ve written about before: a husband and wife couple I visit in Detroit every summer who make their living as business people. They live in one of the few upper-middle-class to upper class enclaves left in Detroit.
They have neighbors down the street, a couple in their late 50s to early 60s. They retired a few years ago as teachers in government schools with pensions of about $60K each, inflation-adjusted. My friends’ only “pensions” are the 401-k’s and IRA’s that they squirreled away in their 6 or so good years out of the last 28 years.
Their neighbors attack the “rich” and the “top 1 percent.” They have my two friends, and some other neighbors, in mind. But the present value of these neighbors’ retirement incomes, if they each live another 30 years and at a real interest rate of 4%, is just over $2 million. If they live another 25 years, it’s almost $1.9 million. Their equity in their house almost certainly is at least $200K. I’ll bet dollars to doughnuts that they don’t refer to themselves as multi-millionaires. They should.
Moreover, as I noted earlier this month, I recently went to a retirement planning seminar, where I learned that if I retire at the end of 2016 at age 66, my annual inflation-adjusted retirement pay will be about $60K. If I live another 25 years after that, which looks plausible, my retirement pay alone will have a present value of over $900K. Combine that with a house in coastal California that’s almost paid for and various IRAs and 401-k accounts, and I should call myself a multi-millionaire.
READER COMMENTS
Silas Barta
May 23 2014 at 8:11pm
How can a few years of work make that big a difference in your pension? If they stop working 3 years before the pension, they literally get none of the pension? Or is there a sharp ramp-up in the last 3 years?
That must create some rather perverse incentives.
Steve Y.
May 24 2014 at 2:26am
One’s “real” economic position, as you wisely infer, should be adjusted by the present values of future unrecorded contractual receipts and obligations, the latter being probably minimal. If the retired teachers made that adjustment to their net worth, they might well be the richest people in the neighborhood.
And who are they getting rich off of? Government entities who likely did not reflect these pension obligations on the balance sheet (I would happily be mistaken).
In their IRAs and 401Ks your private-sector friends likely hold stocks, notes, and cash, all of which are recorded liabilities or equity of various entities. Your friends will draw down these funds in their retirement from entities which have long been providing for these obligations in their financial planning and will not be strained in making them.
On the other hand, pension payments to retired government workers are a major reason the bottom 15% (approximately the official poverty rate) aren’t getting the government services they need. If the teachers want to blame wealthy people for the suffering of the poor, they need only look in the mirror.
Roger Koppl
May 24 2014 at 7:42am
David,
Right. Generally, you should translate everything into one wealth number or one equivalent annual flow, using a risk-adjusted discount rate. The same issue comes up in wealth inequality. Typically, wealth inequality in the US refers to net worth, which leaves some things out, including Social Security.
According to the Social Security Administration, the average monthly benefit for retired workers is $1,294. (http://www.ssa.gov/pressoffice/basicfact.htm) And, “A man reaching age 65 today can expect to live, on average, until age 84.” (http://www.ssa.gov/planners/lifeexpectancy.htm) Using the benchmark interest rate of 5%, that information implies wealth from Social Security of $190,216 upon retirement for the average American man. (Using female life expectancy gives us $201, 646.) Using different assumptions, Forbes magazine has some broadly similar numbers: http://www.forbes.com/sites/advisor/2013/03/22/what-is-the-present-value-of-your-social-security-benefits/. Or see what the Wall Street Journal says: http://online.wsj.com/news/articles/SB10000872396390444592704578064422164043676.
According to the Census Bureau, median household wealth (net worth) in 2011 was $68,828 in 2011 including home equity, and $16, 942 excluding home equity. (http://www.census.gov/people/wealth/files/Wealth%20Highlights%202011.pdf) Now, these census number exclude “equities in pension plans,” which presumably means they are not counting the value of 401k plans. Nevertheless, I think these cobbled-together numbers suggest that excluding the present value of Social Security benefits skews wealth-inequality numbers and tends to exaggerate wealth inequality.
By the way, none of what I have said is a criticism of Social Security. Whether you love Social Security or hate it, omitting the present value of future benefits from wealth calculations likely skews wealth inequality calculations.
Jack pq
May 24 2014 at 10:05am
Prof. Henderson, while I appreciate your point I don’t think it comes across clearly. What are saying, I think, is:
1. For inequality comparisons (if we must), lifetime income matters, not income any given year
2. Lifetime income for a riskless career is much higher, other things equal, than lifetime income for a risky (income wise) career, due to a smaller discount rate (finance 101)
3. Defined benefit plans are worth vastly more than other retirement plans, and this must be factored into lifetime income.
4. The real multi millionaires are not who you think, because of 1-3
David R. Henderson
May 24 2014 at 12:36pm
@Jack pq,
1. For inequality comparisons (if we must), lifetime income matters, not income any given year
Yes, but not just for inequality comparisons, although that is what I do in the post. Also for any real idea of what people’s pay is in a long-term sense, even if you want to look only at those people and not compare them to others. Sorry if that wasn’t clear.
2. Lifetime income for a riskless career is much higher, other things equal, than lifetime income for a risky (income wise) career, due to a smaller discount rate (finance 101)
Lifetime income for an almost riskless stream of income is much higher, all else equal, because of a higher probability of receipt of income for any given year.
3. Defined benefit plans are worth vastly more than other retirement plans, and this must be factored into lifetime income.
Almost. Whether they are worth vastly more or vastly less depends on the magnitudes involved. My point is simpler: that defined benefit plans with inflation adjustment are often simply not factored into the discussion and should be.
4. The real multi millionaires are not who you think, because of 1-3
With all my caveats above, yes.
Patrick R. Sullivan
May 24 2014 at 12:54pm
The implications of all this for Piketty’s analysis of wealth inequality are interesting too.
I’m reminded of one of the funnier episodes of ‘Yes, Prime Minister’, where the PM offers to agree to a pay hike for civil servants if Sir Humphrey Appleby will agree to take a cash buy out of his pension, at Appleby’s (disingenuous) estimate of its value.
Roger Koppl
May 24 2014 at 3:24pm
Jack pq, you are right: The higher probability that you’ll actually get your money is reflected in a lower discount rate.
David R. Henderson
May 24 2014 at 4:28pm
@Roger Koppl,
Jack pq, you are right: The higher probability that you’ll actually get your money is reflected in a lower discount rate.
I don’t think it’s as straightforward as that. Read Chapter 10 of David R. Henderson and Charles L. Hooper, Making Great Decisions in Business and Life.
Vivian Darkbloom
May 25 2014 at 6:34am
Martin Feldstein addressed this succinctly in his recent op-ed piece on Piketty:
“Finally, Mr. Piketty’s use of estate-tax data to explore what he sees as the increasing inequality of wealth is problematic. In part, this is because of changes in estate and gift-tax rules, but more fundamentally because bequeathable assets are only a small part of the wealth that most individuals have for their retirement years. That wealth includes the present actuarial value of Social Security and retiree health benefits, and the income that will flow from employer-provided pensions. If this wealth were taken into account, the measured concentration of wealth would be much less than Mr. Piketty’s numbers imply.”
http://www.nber.org/feldstein/wsj05152014.pdf
Roger Koppl
May 25 2014 at 8:45am
David: Would you mind elaborating? I could not access the whole of chapter 10 of your book with Hooper. You can use a different discount rate for different time periods, as bond specialists do. So PV can handle variation in the riskiness associated with cash flows over time as well as the changes in the risk-free rate. There is the issue of risk aversion, but I guess I don’t see how that factors into our discussion of what a person is “really” being paid.
I don’t think present value is somehow magical and perfect. If novelty is possible, for example, NPV cannot be, like, the One True Decisionmaking Criterion. Your amendment to Jack pq’s statement referred to the “probability of receipt of income for any given year,” so we seem to be assuming a risky world and not an uncertain world. I guess what I’m asking is, “In the more limited context of just saying how much you are getting paid in a risky world, what’s the problem with simply including a (possibly variable) risk premium in the discount rate?”
Norman Pfyster
May 25 2014 at 11:29am
If you combine this analysis with one of Scott Sumner’s favorite dicta that consumption is what matters, pensions have another (usually) unnoticed effect: a lifetime consumption increase. Personal anecdote: my income is quite a bit more than my father ever made. Nonetheless, his lifestyle was considerably more consumptive than mine. Why? He knew he had a pension to cover his retirement, so he could consume almost all of his current income. I, on the other hand, have to save for retirement. For me to build an equivalent savings to his pension, I would need to save approximately $80K per year. If I could spend that $80K, my life would be better, too.
Daublin
May 25 2014 at 11:38am
Silas, there are many problems with pensions. The issue you describe is one reason that 401ks are more logical and intuitive. The payout of a 401k is directly proportional to how long you pay in on them.
Comments are closed.