Wednesday, October 28, 2015

CRFB estimates of solvency effects of Bush Social Security plan

The Committee for a Responsible Federal Budget released their estimates of the effects on Social Security’s financing of the reform plan released yesterday by Gov. Jeb Bush. Overall, the CRFB estimates are very close to the figures contained in an analysis I released yesterday.


What we both find is that the Bush plan could be expected to eliminate Social Security’s 75-year actuarial deficit and to produce a small trust fund surplus at the end of 75 years. Where we differ is that the CRFB analysis finds that, despite the program being solvent over the long term, the Social Security trust fund would go through a period of insolvency that would require that the trust fund be granted the authority to borrow. That borrowing would be repaid by the end of 75 years. In my analysis, the trust fund remained solvent throughout the 75 year period.


The differences are likely due to small details of how we set up the provisions to be simulated. But again, both analyses show a similar picture: that the Bush proposal would substantially improve the long-term solvency of Social Security.


JoeTheEconomist said...

Andrew, I am hoping you can help me understand the estimates on the PPI change. It is looking for 1.45% of payrolls. That is very large. I can't find anything similar to it on SSA's solvency page. Bush claims that it will be phased in over 'decades'. That number simply does not make sense.

Andrew G. Biggs said...

Look in my AEI analysis for more details on the specifics of the provision and the implementation dates. I just ran the provision on its own and found that it reduced the actuarial deficit by 1.64% of payroll. So the CRFB numbers seem good to me.

JoeTheEconomist said...


Just to make sure I understand. If someone had $3,333 in AIME today, they would get .9*($826) + .32*2507 per month roughly $1,572. Under the new plan someone would get .93*($826) + .21*2507 or roughly $1,307.

2,507 is the AIME - the 1st breakpoint.

$3,333 equates to a worker earning about $40,000.

I have to ask a really stupid question. Why don't they simply add in a 4th break-point. Is it the intent here to lower benefits across the wage spectrum?

Today solvency would force a 20% cut in benefits across the board. We are targeting a sub-set of workers decades out. That will mean that we are looking at 20%+ type of reductions on the targeted individuals. Right?


Andrew G. Biggs said...

Looking quickly, those figures seem right.

You could add a 4th bend point, and people have explored that. But I personally see that more for fine-tuning of distributional issues. If you need to reduce costs over time, then using the current formula seems sufficient to me.

WilliamLarsen said...

Now for those who make the average wage which a large fraction of the population and end up with a AIME equal to the average wage index when they turn 60 the benefit would drop by 26.3%. $3,333 per month is 89.1% of the average Wage index. What this will do is cause people to stop working as soon as possible. For those who have saved enough, it may be to their advantage to retire since any increase in their wage will produce little increase in their wage.

Now if you take the theoretical value of the OASI tax over time for this person born in 1948, it would grow for this individual would at age 66 paid $108,243.50 in payroll tax, earned $213,847.32 in US Treasury Interest for a total of $322,090.82. The life expectancy for this cohort is 21.47 years. The theoretical annuity adjusted by inflation over this time is $2,012.03. The initial OASI benefit under current law is $2,072.68 per month. If the Bush plan was fully implemented now it would pay $1640.52 a month, or 82% of current benefits, an 18% benefit cut. The only reason for the theoretical reduction in the annuity calculation is due to the artificially low US Treasury rates. Of course the best investment advice would be to look at paying down high interest rate debt like a mortgage and boost that the 4%, student loan interest of 6%, or take advantage of employer matching 401K contribtions (or all of the above). Effectively loaning SS money at such a low rate while paying higher interest on debt is dumb.

But if we look at the person born after 1985 and assume as SSA that the US Treasury Rate will return to near normal, then we are looking at something a lot different.

This average wage worker pays $331,585.71 in payroll tax, earns $621,910.10 in US Treasury Interest for a total balance at age 67 of $953,495.82 which over the life expectancy of this male 23.51 years of $4,675.03 per month. The initial OASI benefit will $4,081 under the current law and full scheduled benefit of $4,460 under Candidate Bush Plan it would be $3,360. This would pay 72% of benefits of full scheduled benefit, little change from the ability of 2034.

Under current law the person born in 1985 turns 67 in 2052 at which time OASI would be able to pay about 64% of scheduled benefits without COLA or $2,612 per month or 64 cents on the dollar. However the annuity calculation assumes inflation adjustment of 2%. Under current law with no Trust fund, COLA is zero since the ratio of trust fund to expense ratio is less than 20% (final criteria for paying COLA). So adjusting the annuity calculation to pay no COLA, the flat monthly benefit would be $5,656 per month compared to the payable benefit of $2,612 under current law where the individual can expect 46 cents on each dollar of combined payroll tax and US Treasury Rate.

The calculation I did over 25 years ago in the early 90's showed the person born after 1985 to receive 28 cents each dollar of combined payroll tax and US Treasury Rate.

Under Candidate Bush's proposal those making more than the US Average wage will bare the burden of the change while those making slightly less will still loose money compared to paying down debt or investing the same payroll tax in US Treasuries.

To be fair each cohort collectively should not draw anymore or any less than $1 of combined payroll tax and credited US Treasury interest. Some say 75% is not bad, but we are talking the value of a home now.

WilliamLarsen said...

Just one more comment. Does it really make any difference to the layman if his benefit under current law is cut by 20-30% when the trust fund runs out or if a change to the benefit formula is made reducing the individuals current benefit under current law by 20-30%?

No matter how you slice it half the beneficiaries will see the same size cut regardless if nothing is done. Those who created and perpetuated the problem escape the cuts in the future.

Or is it the hope that over time future workers will have forgotten what their benefits would have been had Social Security been actually designed to work. In other words is making this type of change more like a "pain" pill easier to swallow over time instead of actually doing the financially correct solution?

If the future will forever be reduced to a sup par benefit why keep it?