Last week the Congressional Budget Office issued a report on the distribution of major tax expenditures, including the expenditure for “net pension contributions and earnings.” The study was requested by Rep. Chris Van Hollen (D-MD), the ranking member of the House Committee on the Budget.
The initial reaction to the report was predictable — headlines that the tax expenditures primarily benefit the “wealthy.” But as Wall Street Journal online columnist James Taranto noted, the CBO report said no such thing. The report used income quintiles, not wealth, and the top quintile starts with a couple earning at least $115,100 pre-tax. Most couples with that income are probably “comfortable,” depending on where they live and other life circumstances, but hardly “wealthy.”
If we look at the top 1%, the CBO report estimates they get 14% of the employer-based retirement tax incentive. (Contrast that with 68% of the capital gains and dividends tax break that CBO estimates will go to the top 1% in 2013.) But let’s put that 14% number in context. When it comes to employer-based retirement savings, the report is misleading not because of what it says, but because of what it doesn't say.
CBO’s most significant omission is in the discussion surrounding how to view the results and consider the impact of incentives. CBO comments that they allocate the tax incentives to those who directly benefit from them. They also note that some of the savings resulting from the tax incentives is not new savings, but savings moved from non-tax-preferred accounts. But without further comment, both of these remarks are seriously misleading when the subject is the incentive for employer-based retirement savings. So let’s take a closer look.
Employee Contributions Understated
With a tax-preferred employer-based retirement plan, meaningful contributions can only be made for owners and other highly compensated employees by making contributions for other employees. This is unique to the tax incentive for employer-based plans, and is critical to any discussion of the value of these incentives.
When a small business owner decides to put in a workplace plan, or a larger employer puts in a plan to benefit key employees, other employees get employer contributions that they did not have before. Only a small portion of this benefit is recognized in an analysis of the distribution of the tax benefit for these employees, even though this is a very direct benefit, not an indirect impact.
For example, if an employee with a 15% marginal tax rate gets a $2,000 employer contribution, the “benefit" for that employee is really $2,000. What shows up in the expenditure analysis is the marginal income tax rate plus the payroll tax rate times $2,000, or roughly $600, seriously understating the true benefit for that employee. In addition, the owner’s tax benefit is not reduced by the portion of the current year tax savings that is transferred to employees in the form of contributions. The result: The owner’s benefit is overstated and the employee’s is understated. By definition, this could not be reflected in the numbers presented, but it should be mentioned in the discussion.
Impact of Nondiscrimination Rules
The impact of the nondiscrimination rules belongs in the discussion of whether or not new savings is created as well. An employer that “transfers” savings to a tax-favored employer-based account is creating new savings for the employees who now get contributions, regardless of whether or not there is new savings for the owner. This is precisely the result that was contemplated by incentives for employer-provided plans, but it’s not reflected in an analysis of the distribution of the tax incentive.
With regard to the numerical results presented, CBO goes into some detail on how the estimates presented in the report were developed — using a sample of detailed 2006 tax returns and extrapolating them to 2013. Assuming that that information included related W-2 information, those returns would include data on whether or not individuals were covered by an employer plan, as well as elective deferrals made to and distributions made from these plans. This data is not sufficient to determine the tax expenditure for retirement savings.
In order to estimate the tax expenditure for retirement savings, CBO would have to estimate the defined contribution account balances, investment earnings and employer contributions, as well as defined benefit accruals, by marginal tax rate. None of these are reported on federal tax returns. This contrasts with capital gains and dividends, and the deductions and credits included in the report, all of which are reported on individual returns. In other words, the “pension” estimates are not an actual 2006 sample of data extrapolated to 2013; rather, they are estimated 2006 data extrapolated to 2013, and should be viewed through that lens.
An ASPPA report, Distributional Analysis and Pension Tax Provisions, provides a more complete discussion of the challenges in estimating the amount and the incidence of the tax benefit for employer-based defined contribution plans. The analysis estimates that for 2012, more than 70% of the defined contribution tax benefit went to families earning under $150,000. (CBO did not break out DB and DC plans.) Like the CBO analysis, ASPPA’s analysis does not recognize the transfer of the tax benefit from a small business owner to employees. If it did, the benefit would skew more heavily to families with more modest incomes.
The Bottom Line
In summary, the CBO report must be viewed in context. For employer-based plans, the most critical context is the nondiscrimination rules that work to provide significant contributions to millions of workers, and they are not properly reflected in an analysis of tax expenditures. Even without this recognition, the majority of tax benefits go to middle class families. Just think how good it would look if these analyses reflected the true benefit for small business owners and employees.
Judy Miller is ASPPA’s Director of Retirement Plan Policy.
