Retirement Options with $10,000,000

Note:  I’m pretty sure there is an error in the example that I am about to provide, and if so I know my commenters will find it.

Let’s suppose you have $10,000,000 for retirement.  Sounds pretty good, doesn’t it?  How much would that enable one to spend on consumption each year?  Well it depends on your investment choice, but surely one can’t go too far wrong by following the investment advice of a Nobel-winning finance professor:

The Magic Number: 4%

What It Means: This is another spending guideline. It suggests that a retiree spend an inflation-adjusted 4% of his or her total retirement assets each year, keeping the balance invested with a mix of stocks and bonds.

What’s Wrong With It: No less an authority than Nobel laureate William Sharpe, professor of finance, emeritus, at the Stanford Graduate School of Business has written extensively about this “rule” and why it can ultimately be harmful.

The rigidity of the spending plan is among its problems. If a portfolio underperforms, staying the course is a clear path to running out of money. When returns are better than expected, there is an unspent surplus.

Sharpe offers an alternative to the 4% rule. Instead, invest in TIPS (Treasury Inflation-Protected Securities). If those returns prove insufficient, an investor can always dial up portfolio risk and seek better returns.

I can see the advantage of this strategy.  You preserve your nest egg (in real terms), and you live off the interest on the $10,000,000 investment.  So let’s see how it works out.  I will take a fairly simple example, where the money is invested in 10 year TIPS, and the inflation rate is 2%.  10-year TIPS are currently yielding 1.26%, which would provide an annual income of $126,000, indexed to inflation.

But wait, the income is taxable.  So maybe you only have around $90,000 to spend after taxes.  Still that provides a comfortable lifestyle for the average couple.

But wait, I seem to recall reading that the tax on TIPS applies to both the real interest payments and the inflation adjustment.  So let’s assume that there is 2% inflation, which is roughly the recent trend rate of inflation.   Now your “taxable income” will be $326,000, which puts you into a pretty high tax bracket.  Let’s assume your average income tax rate on $326,000 is about 35% (it’d be lower today, but Obama has big plans.).  In that case you’d have a tax liability of $114,900.  Still, it seems like you’d have a lot of money, even after taxes.

But wait, the $200,000 inflation adjustment isn’t really “income,” it is merely a nominal adjustment in the principle invested in your TIPS, which keeps its real value stable.  It’s not like the Treasury sends you a check each year for $326,000; your actual “income” is still just $126,000.  After you pay your $114,900 tax liability, your net after-tax income on the $10,000,000 investment is $11,900/year.  I hope you like dogfood!

Seriously, I know there are all sorts of strategies that can result in higher retirement incomes.  You can take advantage of our finite lifespans by buying an annuity, for instance.  Obviously someone with $10,000,000 would do fine, more than fine.   But I do have a couple serious points to make:

1. From my perspective, the state employees who have pensions with COLAs have a pretty sweet deal.  If I was 62 I’d rather have $65,000 a year for life, adjusted for inflation, than $1,000,000 in a 401k.  Peace of mind.  BTW, the average 55-64 year old has $69,127 in their 401k.  Time to start saving folks!  (Full disclosure, I have a 403b, which is similar to a 401k.)
2. Both wealth inequality and income inequality data are extremely misleading, for all sorts of reasons.  This is just one of those reasons.  I doubt there are too many people who invest in the way I suggested.  Many of the TIPS are probably held in tax deferred accounts, for instance.  But there are definitely some wealthy people who report large interest incomes on their tax forms.  And if they invest conservatively their actual real after-tax income might be only a tenth, or even one one-hundredth of the reported income that shows up in Gini coefficient calculations.  I’m not saying this particular problem is enough to seriously distort those statistics.  It probably isn’t.  But there are also many other flaws that make the indices virtually worthless (flaws which I will explore in future posts.)  In fact, statistics showing consumption inequality are the only valid way to measure economic inequality.  Both income and wealth data are flawed in so many ways as to be nearly worthless.
3. When you read left-wingers complain that workers pay X% tax rates and billionaires only pay Y%, keep in mind that the statistics they are quoting are completely meaningless.  They are assuming that a 35% tax rate actually means 35%, whereas in the example I just gave the effective tax rate was over 90%.  Again, I am not saying this sort of example is common, but even if our tax system were 100% indexed to inflation, the tax rate comparison between labor taxes and capital income taxes would be meaningless, as capital income is double taxation of labor income, and thus the only efficient and fair tax rate on capital income is zero.

Here’s my optimal tax system:

1. A progressive consumption tax or a progressive payroll tax.
2. A set of Pigou taxes (carbon, not cigarettes)

That’s all.