Investigating OASI Privatization

Introduced as part of Franklin Roosevelt’s New Deal, Social Security redefined the role of the government in American life in an almost revolutionary way, similar to the widespread adoption of public education. Despite it’s out-sized impact, Old Age and Survivors Insurance (what most people refer to when they speak of Social Security) continues to be one of if not the most misunderstood government programs.

Before the introduction of Social Security, the few entitlement programs that existed were typically extremely limited in scope and eligibility, and most often run by state and local governments. These programs were typically restricted to the needy, and provided meager benefits. Fast forward eighty years and half of Americans receive direct benefits from the government, with a fifth of the population receiving a monthly benefit from the Social Security Administration.

What I’m interested in is whether the current system is the ideal system for the middle class, and how privatization would work as an alternative.

Before we can delve into the pros and cons of our system, we first have to understand it. Old Age and Survivors’ Insurance (what most people are referring to when they say Social Security) is a defined-benefit scheme. When a person begins drawing Social Security, their primary insurance amount (their benefit) is calculated progressively based on their AIME, or average indexed monthly earnings, which itself is calculated by averaging their indexed earnings over a 35 year period (indexed earnings are earnings indexed to the average wage index for the year two years prior).

With average indexed earnings, the primary insurance amount is determined according to what are typically referred to as the “bend points;” thresholds in a rate-bracket schedule. The primary insurance formula is similar to the formula for federal income tax. Here is the schedule with 2020’s Social Security bend points and replacement rates and how different average indexed earnings levels affect primary insurance amount:

Annualized AIMEReplacement Rate
Up to $11,52092%
$11,520 to $69,42032%
$69,420 to $137,70015%
Over $137,7000%

This progressive defined-benefit scheme is then funded through the Federal Insurance Contributions Act (FICA) with the OASI Tax, which is a flat payroll tax of 12.4% levied on the first $137,700 in payroll for each employee. The tax is split between the employer and employee at 6.2% each. Lastly, primary insurance amounts are adjusted each year to account for changes in the cost of living (COLA).

This scheme as it stands doesn’t necessarily need major reform, but there are a few things to note:

  • The Social Security Trust Fund is projected to be insolvent by 2035
  • OASDI outlays are projected to rise from 5.2% of GDP in 2019 to 6.3% in 2050
  • Savings equal to 12.4% of income invested in stocks and bonds over the last 40 years yielded a significantly greater return than OASI at all income levels

To demonstrate that last point aptly, here’s a chart showing how a full-time low wage worker earning $25,000 a year gets rich with S&P 500 rates of return from 1980 to today (assuming their earnings remain constant):

Despite a very humble income, the worker’s compulsory 12.4% OASI contribution, when invested conservatively in stocks and bonds over a working career, yields the better part of a million dollars. That’s forty times their annual gross earnings, and more than enough for any retiree to live off of, let alone one who (hypothetically) has managed to get by on less than $1,750 a month their whole life.

Further, we see that if the worker were to withdraw $20,000 annually to live on, but keep the rest of their funds invested, they would actually continue to grow their wealth substantially. Regardless of their age of death, their inheritors will be well off.

This is the part where the typical reader asks themselves why Social Security isn’t just a compulsory defined-contribution system instead of a defined-benefit system. There are a few good answers to this question.

There are periods where the stock market does not grow, or where the market is volatile and risky. The S&P 500 did not grow from 1929 to 1954, 1968 to 1978, or 2000 to 2012. Further, it took the better part of a century from the inception of Social Security for stocks and bonds to be easily accessible to the middle class. The answer that takes the cake, however, is the way Social Security was designed.

Let’s say hypothetically you’re trying to repeal and replace OASI. How would you go about doing? If you just simply stop paying benefits and cancel the tax, current retirees who’ve paid the tax for four decades will lose their benefits and many (if not most) will be thrust into poverty. If you phase it out by continuing to pay benefits to those who’ve paid the tax all their life but allow current workers to invest their OASI contribution, you’ll deplete the OASI Trust Fund and explode the deficit because current benefits are funded by current workers. This is a design often referred to as a Ponzi structure of finance.

While there’s no easy way to convert Social Security into a defined-contribution system, it is not impossible, however, it comes with trade offs that ought to be considered. I can appreciate the idea of creating a more equitable system for everyone, but I’m skeptical. As they say, some things are just too good to be true.

Let’s remind ourselves of how it is generally thought that Social Security fails the public:

  • Projected insolvency
  • Inadequate benefits for the needy
  • Low yield compared to the historical stock market
  • Does not produce generational wealth

Now, for the sake of argument, let’s say you were to construct and enact a fairly liberal replacement plan such as the following, which addresses some of those concerns I mentioned up earlier:

  • Replacing the OASI tax on the employee-side with a compulsory IRA contribution
  • Converting OASI into a generous means tested welfare program for seniors
  • Placing OASI on the Treasury General Account when the OASI Trust Fund is depleted
  • Moderately increasing marginal rates of income tax on most income levels

Also, instead of replacing the OASI tax, we could simply have the Federal Government simply manage the OASI Trust like a sovereign wealth fund, which would save money on administrative costs but potentially raise issues of conflict of interest and market inefficiency. In the long-run however, and on the aggregate, both systems should yield the same result, which is the following:

It begins to drop off with the first retirees. I did not factor in retirees who payed partially into the prior system and partially into this system, which means the shape is somewhat misleading. In actuality, the trust balance will reach a smaller level and then decline somewhat more rapidly.
That’s one big sovereign wealth fund.

This chart demonstrates how, using my model, privatization would affect real rates of return on corporate equity, which is interesting:

Finally, below are graphs visualizing the assumptions that were made in order to construct this simulation:

A real projected growth rate of 1.9% till the end of the century. This is extremely speculative.
This was the baseline level of corporate capitalization as a share of national wealth I used, which the additional OASI Trust accumulation was stocked on top of. This is also pretty speculative.
I used the Required Minimum Distribution system for this, though I had it start for retirees aged 67 instead of 70. One flaw of this is life expectancy, which is what the RMD schedule is based on, will undoubtedly increase over the next 80 years, and probably a lot. A higher life expectancy means distributions have to preserve trust solvency.

Finally, the core of this rests on the primitive but seemingly reliable model I presented in my previous article which you can find here. Even if my model is somewhat chaotic and assumptive, what we see in these graphs is pretty enlightening. As we see in this last graph and the trust balance graphs a few graphs above it, the mean distribution (the sum of distributions divided by the sum of recipients) must decline somewhat over time (all things being equal) in order for the trust balance to not be depleted too quickly.

As I mentioned in the caption for the trust balance chart, I did not factor in retirees who began with the previous system making distributions (withdrawing funds from the OASI Trust) on their less accumulated balances. That would flatten the chart somewhat, but what’s more important is that I did not factor in how increased taxes would affect the economy or the inputs to capital accumulation (profits & savings invested).

One of the most interesting things is that the real distribution per retiree in 2100 is only 2.7 times the mean Social Security benefit today despite real GDP being projected to grow by a factor of 4.3 from 2021 to 2100. On the other hand, if you look at real GDP per retiree, you get a figure of about $375,000 in 2021 doubling to to just over $750,000 in 2100. Considering this, the benefit doubling sounds about right but still strikes as unimpressive.

If retirees receive distributions as large as this simulation assumes, they will receive in total in 2100 distributions equal to 32% of gross domestic product, compared to OASDI’s 6.3% projected for the year 2050 (which will supposedly level off for the most part).

Some of the things I think about when considering all of this include how the OASI Trust would likely need to invest in overseas firms in order to make the largest returns in the long-run, and how the United States could in effect take ownership over the world with such an inordinate sum of capital. Another thing I wonder is whether all of this is worthwhile considering in order to do it we must raise revenues elsewhere to fund existing unfunded obligations such as the current OASI benefits to sixty million retirees.

If what we’re concerned about really is the distribution of capital ownership and incomes for retirees, we can achieve the same redistribution through other, simpler means. When we forget the money for a moment, we realize what we’re really trying to manipulate is the distribution of purchasing power of real resources.

Personally I do not have a stance. I can see how a privatization would work, and can imagine how it can be done in a distributionally progressive manner, but I also am satisfied with the current system. It’s quite possible that artificial intelligence and automation will eliminate the need to worry about the distribution of income and capital eighty years from now by, for the most part, eliminating the scarcity of real resources. But who knows!

I hope my readers have enjoyed reading this and have gained some worthwhile insight on the issues of Social Security and OASI privatization.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s