Friday, October 2, 2015

Investing and Insuring Are Not the Same Thing

Wade Pfau's Retirement Researcher blog received a disquieting question from a reader last week. The reader asked Wade how he, the reader, would have done had he invested his FICA payments over the years in the stock market instead of paying into Social Security. We assume the reader knew that wasn't legally possible.

Wade researched the question because well. . . because he's a researcher, and responded along the lines that in the reader's personal situation it would have been roughly a draw, but that investing in the market would have exposed him to far more risk to achieve similar results. In other words, the reader would have done about as well with investments but largely because he had chosen a good time to be born. (He was lucky.)

Wade later noted that his analysis didn't include spousal benefits. Spousal, survival, and other Social Security benefits are huge, so if the reader is married then he would have been worse off investing his FICA taxes. Significantly worse off.

The question is disquieting because it suggests that the reader may not appreciate the difference between insurance, like Social Security Retirement Benefits, and investments. They aren't apples and apples.

Social Security retirement benefits, officially named Old Age and Survivors Insurance (OASI), are insurance against longevity, or living a very long time and running out of wealth. It was passed into law in the 1930's because Americans felt that old people who couldn't work any longer shouldn't live in poverty. We pay the premiums for this insurance with FICA taxes. OASI is available not only to Americans but to anyone who works legally in the U.S. and pays FICA taxes. Benefits are based on how much we pay in FICA taxes over the years.

Insurance is a contract that pays us when bad things happen in our lives but typically pays nothing if we don't suffer a loss. An investment is something we buy with the hope of selling at a higher price in the future.

If you live a long time after retiring, you will be better off buying insurance like fixed annuities or more Social Security benefits, which you can effectively do by delaying claiming. If you live less than an average lifespan, you may well be better off investing – or maybe not, since investments can lose as well as gain. Investing seeks to maximize spending in retirement but guarantees nothing, while insurance guarantees that you will have at least some income no matter how long you live. If you are healthy, you cannot predict how long you will live, which makes it a lot harder to choose between the two. Theirs are two very different promises.

When you buy stocks and bonds you “sell cash and buy risk.” Cash includes things like actual cash and short- to intermediate-term Treasury bonds. Risk means risky assets, like stocks and riskier bonds. When you buy an annuity or increase Social Security longevity insurance by delaying your claiming age, you are buying safety from longevity risk. Another way to say this is that you are paying an insurer to accept your longevity risk.

Your total household balance sheet, as RIIA likes to say (download PDF, should include all social, financial and human capital.

The risk you buy when you invest in stocks and bonds is market risk, or capital market risk, and this is a risk to your financial capital. When you buy an annuity, increase Social Security benefits by delaying claiming age, or buy disability or life insurance, you're protecting your human capital.

Claiming Social Security benefits early and investing them would mean accepting more longevity risk in addition to increasing market risk. Investing FICA taxes in the market would mean increasing your financial risk in hopes of mitigating your human capital risk, which is all perfectly fine if your overall financial situation in retirement calls for more risk. If you have a huge private pension and little savings to invest, for example, that might be the case.

Increasing your Social Security benefits by delaying claiming them would decrease both your financial risk, because you would be investing less, and your longevity risk by making sure that you or a surviving spouse will have more income guaranteed if one of you lives a long time.

When we ask if we would be better off with stock investments than Social Security benefits, we're asking if it is better to gamble that our investments will do well and we won't live a long time or to buy insurance that will increase our income if we do live a long time. The ideal answer is to have enough wealth to do some of each. This is the basis for a floor-and-upside income strategy.

When I read the question, my first thought was why not invest my homeowners insurance premiums in the market instead of handing them over to an insurance company? If my investments do well and my house never floods or burns down, I'll be way ahead of an insurance policy. (Please don't try this at home.)

But what if it does? What if my house burns down? What if I live to an old age? What if  my widow lives to a very old age? Some risks, like long-term care costs, home insurance, and living to 95 are, as my cousin in the insurance industry likes to say, not easy to out-save. When there is a low-probability, high-magnitude risk and insurance is affordable to protect against it, then insurance is the most effective way to deal with that risk.

The next question might be, “I’m wealthy and don’t need Social Security, so why should I have to participate?”

The answer is adverse selection. If only people who really need insurance buy insurance, any kind of insurance, that insurance is doomed to failure. Successful insurance only works when claims are random and uncommon.

Adverse selection increases the price of life annuities (SPIAs and DIAs), as well. People who don’t expect to live long buy life insurance, not annuities. Those who buy annuities are those who think they will live longer, so the price goes up. If only poor people or old people enrolled in Social Security, it couldn’t work.

So, why am I bothered by someone asking if investing is better than insuring? Because they are not the same thing, so “better” depends on what you are trying to achieve. Is an umbrella better than sunscreen? It is if it rains.

Investing and buying insurance are two very different propositions. Even if your personal preference is to bet everything on the market, you should do so understanding this difference.

A bit of housekeeping. The URL "www.theretirementcafe" recently became available and now brings you here! Please bookmark it.


  1. Is the Total Household Balance Sheet available as a download? The overview provides excellant information.

    1. Clicking the "download PDF" link in the post doesn't work for you?

    2. No, it only gave an overview page, with no link to what is presumably a spreadsheet one can fill in.

    3. Ah, I'm not sure the spreadsheets are publicly available. I assume they're intended for financial advisers with the RMA designation.

  2. Thanks. I agree that such comparisons are very difficult. However, I would have also focused on a few other major considerations to address in such a comparison:
    First, higher income Americans would not purchase such "insurance" where the benefit formula is so regressive as to income. You make a comparable point when you discuss the purchase of annuities when individuals consider their life expectancy. People complain all the time that FICA taxes (the 6.2% for OASDI) are regressive as to wage incomes because of the cap on FICA wages (and that FICA taxes only apply to wages). However, the reality is that the benefits provided under OASDI are much, much, much more regressive to income compared to the taxes used to fund those payments. That results from the bend points in the benefit formula. So, there is a wide variance in the results based on earnings over the 35 year averaging period.
    Second, you ignore the legacy benefit for the substantial number of Americans who are not married when they die, or who do not have an unemployed surviving spouse > age 60 or minor age dependents when they die. You claim such survivor benefits are huge, but, given the dramatic increase in two-earner households, the difference between an individual's own benefit and the surviving spouse benefit is likely to be less than decades ago when we had many more nuclear families.
    Third, you ignore the wide disparity in life expectancy among different races, ethnicities, genders, and even within those categories - as studies show the nature of work, the household income level and other factors have a significant impact on longevity.
    Fourth, you mention being born at an opportune time. Agreed. The greatest generation achieved a favorable return on their FICA taxes, the boomers generally will have a very low to negative return on their FICA taxes, while Millenials will probably end up losing big time. See: I believe the rates of return under the increased payroll tax and reduced benefits scenarios are significantly understated (because the status quo cannot be maintained once trust funds are exhausted). Another way of saying this is that to maintain the benefits, you likely shift more debt to future generations - just like actions taken in the past). And, of course, the Social Security estimates don't include the impact of taxes on those benefit payouts.

    I certainly agree that comparing a lifetime of saving with social security benefits is a very difficult comparison. However, I would have simply referred the individual to the SSA calculations and noted that because Social Security is mandatory, any such estimate is an exercise in futility. I think the calculation is certainly different for any individual as a review of what happened (after retirement, or after death once benefits cease) - versus calculations that incorporate projections. However, we will soon get significant experience in the perils of projections once the DOL issues regulations requiring plan sponsors project income streams for individual account retirement savings plans. My experience is that unless you use purchase rates for actual, immediate annuities, such projections are more likely to mislead than to inform. However, the damage is likely to be minimal, as few read mandatory disclosures anyway.

  3. Jack, a couple of points.

    First, my post is not an analysis of Social Security benefits. It is an argument that insurance (Social Security, life insurance, disability insurance, life annuities) shouldn't be compared with investments using the criteria of return on investment. They are two different transactions. Hence, the title of the post. I don't believe I "ignored" anything in my analysis, because I didn't do an analysis.

    Second, while I agree with you that comparing purchasing investments to paying FICA taxes is legally pointless, comparing investments to the other forms of insurance is not a legal issue, but a financial one. And that's what my post says.

    Thanks for writing!

  4. Dirk, what are your thoughts on buying Long Term care insurance considering the history of adverse selection, insurance companies increasing premiums and difficulties in obtaining payments when needed?

    1. John, I think there are no good choices at present for many households and that this is a "lesser of two evils" decision. I have a client whose premiums just went up significantly, but can continue to pay them – painfully – at present.

      A friend in the insurance industry is convinced that this insurance doesn't work for insurers or their clients and that it will eventually not be offered.

      David Blanchett published some research on retirement spending in which he found that households that don't under-spend or over-spend tend to see their expenses decline 1% to 2% annually throughout retirement in real dollars. More relevant to your question, he found that even for households with high end-of-life costs, those costs are lower in real dollars than early-retirement spending.

      Still, catastrophic health care spending is a possibility and we should have a plan for it. (Only a small percentage of these costs are catastrophic. Many stays are brief and can be paid out of pocket. See Long Term Care Insurance, a post I did some years back.)

      I think there are three tiers of households. Many simply can't afford the premiums. Wealthy households (I hear numbers like savings of $2M per spouse) can probably self-insure. Those households in between are the ones who need to make a decision and face the risks you mention plus the risk that their insurer will get out of the business.

      Joe Tomlinson is the retirement expert whom I most trust on insurance issues and he has stated that those in the middle tier, and perhaps the wealthy tier, are probably better off buying the insurance despite all the known issues with LTC insurance. Probably because it is the lesser of two evils.

      I wish I had a better answer for you. Unfortunately, I don't think there is one.

      Thanks for writing, John.

  5. A better - and simpler - analysis of SS RIB is to look at the payments just to OASI and the benefits just for RIB (and SIB in the case of married couples).
    I actually set up a spread sheet, input the OASI tax rates, the salary index data, my income history, and veterans benefit. I (and my employer) paid a total of 87,000 dollars between 1968 and 2000 (when I determined that I had just worked enough). Even adjusted for inflation (using the salary index as a proxy for CPI inflation) my total contribution in current dollars is just 174K.

    My AIME is $4000 and PIA is 1740.

    If 1740/mo indexed for inflation is worth (using a 4% rule) 520K, I would need to have had a return of 7.08%. If I use a 5% initial indexed payout my return was 6.2%. If I bought an annuity today (not inflation indexed) with $175,000 I would probably begin receiving less than 1000/month.

    SS has been verr good to me. And this analysis does not include survivor benefits for my wife.

    1. That sounds about right.

      It's quite unfair to Social Security in the analysis to exclude spousal and survivors benefits where applicable, which would make it an even better deal in your case.

      A commenter above mentioned that spousal and survival benefits aren't as important now because there are more two-earner families. That is true in the broad sense that there are now more spouses whose own benefits are greater than their spousal and survivors benefits would be, but that is irrelevant to millions of Americans who still claim spousal and survivors benefits. For those households, it can be a very big deal.

      Let me once more point out that, although I'm happy to talk about any retirement finance issue any time, my original post was not an analysis of Social Security benefits, but an argument that we should not compare insurance and investments on the basis of rate of return. Social Security retirement benefits and other annuities will guarantee income for no matter how long you live. Investments only offer a possibility of doing that.

      Thanks for sharing!