Prepare for the 50 year Treasury bond …

Since our newly inaugurated President seems intent on partial privatization of Social Security (despite lukewarm support from his own party), I thought I should offer him a bit of somewhat constructive advice.

Fund your plan by issuing 50 year bonds.

After all, long-term investments should be funded with long-term debt.

Is the partial privatization of Social Security -- something along the lines of the Commission’s plan 2, even though Dick Cheney is right in that it keeps costs down by only providing puny private accounts, not manly Texas-sized private accounts -- the kind of thing that should be financed with ten year bonds? I am afraid not. The plan moves forward the date when Social Security benefits exceed payroll tax revenues from 2018 to 2006. Do nothing and in 2015, an unreformed Social Security system would still be lending a (small) surplus to the rest of the government; adopt the Commission’s plan 2, and Social Security will be paying out more than it takes in.

What about twenty year bonds? Alas, The scale of the net transfer of general funds required under the Commission’s plan 2 is larger in 2025 -- even with the benefit cuts -- than it would be if we did nothing (1.03% of GDP v 0.64% of GDP). That’s right: the cash flow pressures placed on the rest of the government for the first twenty years of the plan 2 are unambiguously bigger than the cash flow pressures associated with doing nothing.

Maybe we should revive the thirty year bond? Or the other hand, maybe not. The Commission’s plan 2 would exhaust the Trust Fund in 2036, not 2052. A maturing thirty year bond would come due just as the government needs to be gearing up to provide the transfers of general revenue the Commission proposes to keep Social Security cash flow solvent between 2037 and 2050.

Don’t believe me? Let’s see what the CBO says:

CSSS Plan 2 would enable the government to pay the benefits scheduled under that law without transferring additional money from the general fund until 2036. From 2036 through 2050, transfers from the general fund would be required.

These transfers are a form of cheating. The Trust Fund -- past payroll tax surpluses -- is already depleted at this stage with plan 2. But benefits still exceed revenues, since the shift to inflation-indexing is gradually phased-in. The Commission squares this circle by just giving Social Security the money it needs. Give money to the current Social Security system, and it would be solvent after 2052 too.

Maybe 40 year bonds? Alas, Social Security benefits still exceed payroll tax revenues at that time. That’s right. Under the proposed reform, Social Security benefits, even with large benefit cuts for future retirees, would exceed payroll tax contributions until 2050.

To be totally fair, by 2045 Social Security benefits have been cut to the point where the net transfers from the rest of the government under Plan 2 are smaller than the transfers associated with keeping the status quo (0.43% of GDP v 1.23% of GDP). There is a difference, though. With the current system, the rest of the government is paying the Social Security Trust Fund back for past payroll tax surpluses it borrowed: the CBO says the Trust Fund will last til 2052 without any reforms. Under Plan 2, the transfers at this stage are just a gift.

But with plan 2, there will be small cash flow savings by 2050, and huge savings by 2075. HUGE. Guaranteed (or maybe not). Sounds to be like the kind of proposal that should be funded with a 50 year bond. That would match the costs (years 1-45) to the payoff (46 on) -- so long as we keep payroll taxes higher than benefits after 2050 ...

The bottom line: The Commission’s plan 2 eliminates cash flow deficits after 2052 (the source of the unfunded Social Security liabilities that so worries Secretary Snow) by creating cash flow deficits in Social Security from now until 2050 ...

No way any private investor would fund a project with such an uncertain long-term payoff -- at least not without a government guarantee. Imagine the prospectus. We have no clue what the world will look like in 2055, but it is safe to assume it won’t look much like what we say it will look like ...

By the way, I love how de facto Treasury Secretary Dick Cheney "worries" about the risk of not cutting taxes enough, or about the risk of not creating large enough private accounts, but he shows no sign of "worrying" about the risks of borrowing too much. Personally, I like Treasury Secretaries who recognize that deficits do matter, and who understand that funds borrowed by the government are funds that are not available for private investment.

I would love to see the CBO’s analysis of the impact of the "Cheney deficits really don’t matter social security privatization plan" -- diverting up to 6 percentage points of Social Security revenue into private accounts (with no cap, one assumes) would radically increase the budget costs of the reform. It takes between about 1/2 of the program’s revenues away, and barring a commmensurate cut in near term benefits, implies an enormous need for new borrowing. Plan 2 -- which caps contributions to private accounts at $1000 a year, limits the diversion to about 1/6 of program revenue. Sounds like Cheney wants to float lots of 100 year bonds. (note: small edits to last paragraph)