Thursday, September 4, 2008

How will Social Security progressivity change under Obama plan?

A blog is a good outlet for small pieces of information that might not otherwise see the light of day. Here's a good example of that. I was interested in how the Social Security tax burden would change under Sen. Obama's proposal, which I'm here taking to be a 4% surtax on individual earnings over $250,000, with no extra benefits paid. (Note that the plan itself is incomplete, so this is a partial analysis of what an overall reform package would look like.)

There are a number of ways to analyze progressivity, but the approach here may be easy to understand. Using the Policy Simulation Group's models, I simulated the Obama proposal for the 1990 birth cohort, so that they would live almost their entire working lives under the new plan. I divide the population into ten deciles based on their level of lifetime earnings. For each decile, I measure the percentage of total taxes they pay and the percentage of total benefits they receive. Low earners pay a much smaller percentage of taxes than benefits they receive, due to the progressivity of the benefit formula. At the high end, the pattern is reversed.

The Obama proposal would affect only high earning individuals. While a very small percentage of individuals earn over $250,000 per year, they are not the same individuals in each year so a somewhat larger percentage of individuals would be affected. Nevertheless, it's really in the top 10% of the earnings distribution where the effects are felt.

Distribution of total tax payments and benefit receipts, 1990 birth cohort


Obama plan

Current law

Lifetime earnings decile

Percent taxes paid

Percent benefits received

Percent taxes paid

Percent benefits received

10%

1.5%

5.7%

1.6%

5.7%

20%

2.9%

6.5%

3.0%

6.5%

30%

4.0%

7.2%

4.1%

7.2%

40%

5.2%

7.8%

5.3%

7.8%

50%

6.4%

8.7%

6.5%

8.7%

60%

7.9%

10.0%

8.1%

10.0%

70%

9.8%

11.4%

10.1%

11.4%

80%

12.2%

12.8%

12.5%

12.8%

90%

15.0%

13.9%

15.4%

13.9%

100%

19.4%

15.0%

18.2%

15.0%

Under current law, the top 10% of lifetime earners pay around 18.2% of total taxes, but receive only around 15.0% of total benefits. Again, this is due to the progressivity of the benefit formula. Under the Obama proposal benefits would not change, so the top 10% would continue to receive 15.0% of total benefits. However, their share of the total tax burden would rise to 19.4%. Again, note that the Obama proposal would fix only around 15% of the long-term deficit, so other changes would be needed in addition to those shown here.

3 comments:

shoffy22 said...

Very cool distributional info!

Comparing the percent taxes paid under the Obama plan to current law, I wonder if these results show that a provision like this could help us tackle some conflicting economic challenges:

1. To promote saving and investment, there seems to be a pretty strong consensus that tax rates on investment (capital gains, etc.) need to be relatively low.

2. Our current income inequality has grown to a degree that those in the top 10% (and especially top 1%) of lifetime earnings have a much larger share of total earnings than in the past.

3. Since a larger portion of the income of those in the top 10% of lifetime earnings comes through investment income that is taxed at lower rates than wages, I think the highest 10% of earners are paying a lower share of total taxes, when measured as a ratio of their share of total earnings, than in the past.

4. To compensate for this, I think the share of total taxes paid by moderately high income people who still receive most of their income through wages has gone up.

5. The results here look to show that from this provision, those in the top 10% of lifetime earnings would pay a greater share of taxes, and also that those in the 60th through 90th percentiles would experience the biggest reductions in their shares of taxes paid.

Thus, it looks like a plan like this to have a surtax on very high earnings above $250,000 could help achieve the goal of making the overall tax burden distribution more reflective of our widened economic inequality, and also allow us to maintain low taxes on savings and investment. This might be a pretty cool result that has pretty widespread support across the political spectrum perhaps?

Andrew G. Biggs said...

Dear Mr Shoffy,

Good points! A couple thoughts. Yes, inequality has grown, although Richard Burkhauser has a new NBER paper that argues that much of the increase since the early 1990s was a product of how Census handled top-coding of high incomes.

Re point 3, you can check out some data here: http://www.cbo.gov/doc.cfm?index=8885. It appears that the top 10%/1%/0.1% are paying record proportions of total federal taxes (see Table 1B).

I guess the bigger question is whether, assuming you want the tax code to equalize incomes, Social Security is the appropriate tool to do so. It's not been used in that capacity in the past and it's arguable that doing so would undermine the 'earned benefits' aspect of the system. The income tax code may be a better tool for the job.

shoffy22 said...

Good thoughts Andrew! And the data from the CBO link is pretty fascinating!