Tuesday, February 23, 2016

Social Security Benefits in a Lake Wobegon World

There’s been an active debate over how to measure Social Security “replacement rates,” which represent retirement benefits as a percentage of pre-retirement earnings. I revisit the debate, and add some new numbers, over at Forbes.

The takeaway: some new figures from the CBO make Social Security benefits appear small relative to pre-retirement earnings. But that’s because the CBO calculations don’t count all Social Security benefits and they don’t count all pre-retirement earnings.

lake wobegon


WilliamLarsen said...

Replacement rates for Social Security OASI is a red herring. The way the CBO, SSA and others are looking at it is incorrect. The best measure and the only true measure is to view it in terms of what the SS-OASI payroll tax paid by a workers would pay invested in identical US Treasury Rates.

Some would call this an accrued basis calculation. What the CBO, SSA and others fail to see is the time value of the funds paid to SS-OASI. In simple terms each cohort should take no more out of SS-OASI as a group than they contributed.

If you look at a person who made $30,000 wage each year starting from age 22 and working to age 67 and assume their wages stayed up with the SSA Wage Growth one can easily estimate the value of their contributed payroll tax to SS-OASI. Using SS Cohort life tables we can calculate the theoretical annuity indexed by cpi just as SS-OASI is scheduled to do. Assuming 1.5% wage growth and 1.5% cpi and 4% US Treasury rate this workers would have at age 67 $494,422. It would pay $15,809 in today's buying dollars which would be indexed by wage growth rising to an initial annuity of $30,894 a year.

If Wage growth were greater say, 2% and everything else the same, the fund would grow to $541,128 and the annuity would be $13,879 in today's buying dollars and would grow to an initial annuity of $33,813 a year.

Now if you noticed, an increase in wage growth without changing any other variable actually reduced the initial annuity, yet increased the annuity at age 67. This is due to the annuity being based on the average wage increase each year.

In this assumption the person made $30,000 a year and theoretically the annuity was $15,809 or 53% of wages at 1.5%. When you look at last years wage at age 66 and the annuity it is still 53%.

At 2% wage growth the person made $30,000 in their first year and the annuity was $13,879 or equal to 47% of wages.

Now many people do not work continuous so the amount would be less.

Targeting a replacement rate is just plain wrong based on the 1977 SS-OASI benefit formula. This formula does not adjust the payroll tax based on wage growth. This by far is the single largest contributor to the current problem. Changes in cpi and US Treasury rates make a difference. Since 2008 treasury rates have been artificially low so even this causes problems.

In short, targeting a replacement rate of 50% maybe high if one were to stay with US Treasuries.

There are other ways to obtain a higher rate of return over all instead of going to treasuries. High interest debt while contributing to payroll tax is anti wealth creation for the working class. Borrow money to buy a house, college that either reduce long term costs therefor increasing savings or investing to get a skill set that pays better is far better than the payroll tax.

So what is the purpose of Social Security? If it is a safety net, then why should I be eligible? Sure I paid into it since 1973, but if it is a safety net, I do not need it. I would much rather pay a much lower payroll tax and target those who need it then to pay everyone something. It is using the shot gun approach to save 10%.

Replacement rate calculations only show how terrible social security is.

Arne said...

Replacement rates as redefined by financial advisors applies primarily to those people who are trying to decide whether they can afford to retire next year - clearly a Lake Wobegon group already by virtue of being willing to hire a financial advisor.

To apply it to everyone, you need to define it clearly, but you should realize that in doing so, you are redefining it.

Andrew G. Biggs said...

Arne, I disagree. The idea of a replacement rate is to measure how well your retirement income allows you to maintain your preretirement standard of living. For all but the very poor, that's the goal of saving for retirement. Earnings aren't the same as consumption, so we can't measure your standard of living directly. And most people don't consume their whole income, especially in the years just prior to retirement. But your average earnings over time should give a decent indicator of what your consumption level will be.