Monday, March 2, 2015

Social Security projections: Why do SSA and CBO differ?

Over at Investor’s Business Daily, Jed Graham looks at why CBO’s projections for Social Security’s financing are more gloomy than those from the Social Security Trustees. One issue Jed doesn’t highlight, but which I think plays a pretty big part, is that CBO takes a more optimistic view regarding longevity improvements than SSA does. Lower mortality is great for us, but not so great for Social Security.

Jed does hit on one unusual point:

CBO expects ObamaCare to shrink employment by the equivalent of 2.5 million full-time workers a decade from now — reducing Social Security payroll tax revenue. But SSA doesn't see any negative impact on work or earnings.

SSA Chief Actuary Stephen Goss told IBD that his department looked at Medicare's 1966 launch and "did not see any big move in the (labor force) participation rate."

I find the response from SSA confusing: CBO projects lower employment due to the ACA based on a number of very specific parts of the law that lower the return to work and, in some cases, can increase an individual’s income if they work less. While Medicare also provided people with greater access to healthcare, in the vast majority of cases these new beneficiaries were age 65 and up, making comparisons to the ACA dubious, to my mind. I can say for sure that the CBO’s projections for labor supply are correct – though I suspect they are – but SSA should analyze the work responses to the ACA in the same context that the CBO did, not simply by looking back at historical examples that may not be very relevant.


WilliamLarsen said...

Labor participation rates are way down. It is doubtful that the US will reach the participation rate of just ten years ago. This will little to no affect on future OASI benefits, but will have a very large affect on OASI revenues. We should be at around 170 - 173 million workers, when we are at 160 million. Will OASI revenues drop by 8 to 10% from their projections by SSA?

JoeTheEconomist said...

Andrew, The key impact to me of the ACA is wages. It is a confusing effect because it pulls in multiple directions.

In some cases, wages will rise as employers cease to offer health insurance so that employees can capture the ACA subsidy. This will have the effect of increasing wage levels.

On the other hand, employers who are forced to pay higher premium(As I Am)will lower what we are willing to pay in wages to employees.

WilliamLarsen said...

Any forecast of revenues and benefits paid to beneficiaries when it comes to OASI has little to do with wages paid. Wages are indexed by the change in the US average wage thus increasing or decreasing the initial OASI benefit. Wages also determine OASI revenues. The small amount paid to the OASI trust fund is a minimal in terms of total liabilities being offset.

If wage growth slows or decreases there is a corresponding drop in the initial OASI benefit. If wages increase so do revenues as well as the initial benefit.

Is there really any impact from the ACA in terms of slowing or cutting wages, no?

Both the CBO and SSA need to stop looking at GDP and look at actual values such as wages and workers.

I think this is a red herring.

JoeTheEconomist said...

Bill, I am making an assessment for revenue for the system. As wages fall, the revenue to the system will fall. While initial benefits are lower, the cost of existing benefits continues to rise with COLAs. I do not think that will end well.

Andrew G. Biggs said...

A decline in wages will hurt the system's finances, even though benefits eventually adjust. The sensitivity of OASDI's finances to wage growth isn't as big as some think -- roughly, a 1 percent change in annual wage growth leads to about a 1 percentage point change in the actuarial balance -- but it does exist.

WilliamLarsen said...

Once a cohorts initial OASI benefit is set (determined by wage indexing), the only impact is COLA (+/-) For example those who retired in 1975-1976 actually ended up with higher OASI benefits than those who retired in 1977,1978 and 1979. COLA in those years boosted current benefits far faster than wages. Of course the notch babies would influence the actual benefits paid.

Because the average life span of those at full retirement age is around 19 year, we would expect the full impact of slowed or decreased wages to take 20 years to truly be felt.

In a fully funded program, the ups and downs of wage indexing (change in wages (+/-) would not be a problem. However, because the OASI trust fund is insignificant in size to the total current liability, it can provide no protection from economic growth and recession.

People need to face the reality that 2029 or 2033 or what ever year OASI faces the inability to pay scheduled benefits that this is a really short period of time away.

A worker who goes to work at age 21 and reaches full retirement age at 67 has 46 years to accumulate sufficient assets to provide for their retirement. This 46 years is twice the number of years the OASI program has before it can not pay scheduled benefits under current law. Now add in the fact that we have 46 cohorts with work histories that make matters worse. To top it off there are more than 40 cohorts that are drawing benefits.

In terms of scope is this really matter at all?

Is this really any different than those on the SS Titanic faced?