The stock market can’t save social security

Good morning. Caroline Ellison, one of the leaders of defunct crypto exchange FTX, just got sentenced to two years in prison. Her former paramour and boss Sam Bankman-Fried is now reportedly bunking with P-Diddy. What celebrity should Ellison share a cell with? Email us with your choice: [email protected] and [email protected].

A sovereign wealth fund for social security?

Both the Trump campaign and the Biden-Harris administration floated the notion of a US sovereign wealth fund a few weeks ago. This is generally a bad idea, for reasons that various people have pointed out. There is, however, one version of the idea that is at least intriguing.

For a sovereign wealth fund, one needs sovereign wealth — specifically surplus sovereign wealth. The US doesn’t have any. No large pools of excess foreign exchange reserves, like Singapore and China; no massive cash stream from natural gas and oil, like Norway and Saudi Arabia. The US government runs a massive deficit, and what oil money there is already goes to reduce it.

Lacking a surplus, other ways to raise a fund — issue debt, new taxes or tariffs — mostly just shift money around. The money that the government raises with taxes or borrowing would otherwise have been invested or spent somewhere else, and there would only be a net benefit if the government invested it in a more growth-friendly or strategic way than it would have been otherwise. Add to that political hurdles (Congress’s attitude towards a pile of money that sits outside its authority is easy to imagine), redundancy (federal programmes already target the sectors that a fund might support) and market impacts (crowding out, asset inflation). 

But there is one pool of American money that could benefit from higher returns. Social security, the US government pension plan for citizens, has a poor financial outlook. The payroll tax revenue that funds the plan once exceeded the plan’s outlays, and the excess cash was invested in Treasuries. But starting in 2021, outlays began to exceed inflows, and the Social Security Administration started to draw down the saved funds to make up the difference. The Congressional Budget Office projects that the reserves, currently around $2.7 trillion, will run out in 10 years.

Benefits will still be paid after the funds run out. But without a new source of funds, the total benefits paid to retirees would be 23 per cent lower at first, and would continue to decline as the US population ages. Republican Senator Bill Cassidy and independent Senator Angus King (and a motley crew of pundits and economists) have floated the idea of reinvesting the trust funds in the market to get a higher return. Now take the idea further: the US government could borrow at the long-term Treasury rate (4 per cent or so), and invest in US equities (6-7 per cent long-term average returns), with the proceeds of the arbitrage going to social security’s reserve funds. 

This would allow the US government to, in effect, create a sovereign wealth fund for a single, well-defined purpose: better returns for a better-funded pension plan. No strategic investments in emerging sectors of the sort Harris and Trump envisage. Just an arbitrage. 

Would it work? We made very simple models of three scenarios: (1) investing the current reserves in the market (not an arbitrage, per se); (2) investing the reserves and an additional $1.5tn dollars, raised through a bond issue; and (3) investing the current funds and borrowing whatever was needed to fund social security though 2055. We made some assumptions:

• We assumed funds invested in US markets return the historically normal 6.9 per cent a year, ignoring the (very real) possibility of wide variation around that mean over multiyear periods. We ignored, in other words, the political and financial repercussions of a possible market crash. 

• We assumed that government equity purchases would be tax-free when sold.

• We ignored the possibility that a massive new Treasury issue might drive yields down from current levels, and the possibility that a multitrillion government investment in US stock markets might inflate asset values and dilute returns.  

• We made the simplifying assumption that new money enters the fund at the beginning of the year and benefits are all paid out at the end.

On these assumptions, investing the current $2.7tn in reserves in stocks rather than Treasuries would extend the life of the reserves to 2040 — six additional years. On to scenario 2: the current reserves are topped up with $1.5tn in funds raised today at the current 30-year Treasury yield of just over 4 per cent, and new funds are left to compound until 2040, when they begin to be disbursed. This massive cash infusion stretches the fund only until 2046. That’s helpful, and makes use of the arbitrage between US government borrowing rates and average market returns. But that still only gives social security an extra 12 years from the most recent CBO projection, while growing the US debt by about 4 per cent and (for one year) doubling the annual budget deficit. 

For scenario 3, we calculated the amount of money that would fund social security in full by 2055. To do that, the US government would need to borrow and invest a little less than $2.4tn today.

Even under our extremely charitable assumptions, the extra returns offered by stock markets only do so much to solve social security’s funding problems. A massive infusion of cash is still required. The extra return would help, but would bring with it all the very real risks we have ignored in our toy model, political controversy, market distortion and financial volatility first among them.  

(Reiter and Armstrong)

One good read

(Not) Made in America.

FT Unhedged podcast

Can’t get enough of Unhedged? Listen to our new podcast, for a 15-minute dive into the latest markets news and financial headlines, twice a week. Catch up on past editions of the newsletter here.