Should the IRS get involved in impact investing?
It’s a clickbait-y title; I’m not suggesting the IRS suddenly start investing taxpayer money into impact investments using the US Treasury balance sheet, whether they be impact funds or impact bonds (etc.) However, maybe the IRS should start offering programs that encourage impact investments so that federal government can save money in the long run. It’s not as if such programs are unprecedented — just look at the IRS Opportunity Zone tax programs.
I recently wrote about why the Fed should consider backstopping impact investments for financial sponsors to encourage investment in impact assets when investment in such assets wouldn’t otherwise be an explicitly preferred choice. You can find the article here. The proposition is an extremely low-cost solution for encouraging impact investment (essentially by providing insurance for investments in impact-generating firms). The benefit of this proposal is twofold between the likelihood of realization and the ultimate results: First, legislating to have a quasi-government entity (e.g., the Fed) provide an insurance-like product would likely be more straightforward to accomplish than it would be to implement other programs to encourage impact investing, such as a direct subsidy; second, the program would promote impact investing in situations when impact investing is not the obvious choice, which consequentially would expand the asset class.
Another way to encourage impact investing and thereby expand the relevant share of investments made into impactful companies — without seeking a direct subsidy — would be to offer tax benefits. The relevant incentive provider would be the IRS, and unlike the insurance-like product of backstopping investment losses incurred while investing in companies verified as impact-generators, there would be a direct cost to offering this program: foregone revenue.
The potential for abuse of the program would need careful consideration when legislating the impact tax incentive program. The program would need to be structured so that any tax benefit received due to the company’s status as an impact-generator must be reinvested back into the business rather than paid out as a dividend to shareholders. This tax benefit restriction would effectively guarantee that the government would be paying for impact, rather than providing a bonus to equity owners.
The key to implementing such a program would be to measure long-term gains from foregone tax revenue now. Someone legislating such a tax incentive program would need to be able to point to the impact verification industry (when this industry exists sometime in the indefinite future) to establish the framework that impact can be reliably quantified and that a professional class dedicated to forecasting impact exists (to whatever degree of reliability can be achieved). In conjunction with the products of the impact verification industry, you could use both case studies and aggregate measures of impact in the market to demonstrate that the program would effectively pay for itself.
For example, consider an entirely hypothetical company that sells educational products and services to prisons, allowing convicts to receive an education superior to what they otherwise would get by reading prison library books. Superior education during incarceration would allow these men and women to, upon release, obtain middle and high skill jobs across blue and white-collar professions, rather than just low-skill, entry-level jobs for the rest of their lives. For the sake of argument, let’s say this company has an impact multiple of money (IMM) of 25%. In other words, for every $1 invested in the firm, the firm yields social impact at a rate of 25% when translating the social value into dollar terms (e.g., $1.25 of social impact when measured the next year, $1.56 of social impact the year after that, and so on).
A company such as the one described above is well-suited for impact measurement because the societal value can be calculated in a relatively straight forward way. You could consider recidivism rates for inmates who do not use the firm’s educational products and services (control population) and those that do use the firm’s educational products and services (target population). From there, include average sentencing lengths and any other relevant factors to “fine-tune” the calculation to arrive at the total taxpayer burden that could be alleviated from the firm’s widespread operations.
To calculate a “rate of return” on the foregone tax revenue from the IRS, you would have to understand the amount the firm’s operations could expand under an impact tax incentive program. Each impact-generating firm participating in the impact tax incentive program would save the amount of money equal to the below equation:
Pre-tax income * effective tax rate (all firms) — pre-tax income * effective impact-generator tax rate
Because impact generating companies would have a lower effective tax rate (I propose we call it the “effective impact-generator tax rate”, although maybe someone can come up with something more catchy), they would be able to reinvest into their businesses the amount of income saved by paying the effective impact-generator tax rate. The social return on the expanded operations would hopefully show a return on capital that would be higher than the value of the foregone revenue that the IRS would have collected had there been no impact tax incentive program to begin with.
For example, imagine that the lost revenue from an impact tax incentive program (let’s call it $100) would have otherwise been used to fund prisons. In the short term, prisons would experience a budget shortfall of $100, which the government would likely need to make up with lending (let’s assume the government borrows at an annual rate of 5%). Over a 5 year period, that $100 costs the government $125 ($100 principal + $5 annual payment x 5 payment periods).
However, this might actually be a good deal for the government! Let’s say the IMM measurement for the prison education firm exclusively represents value derived from recidivism rates. If the firm receives $100 (e.g., the value derived from the impact tax incentive program), it can produce ~$305 in social impact from the reduced cost of housing fewer inmates due to lower recidivism rates [$305 = ($100 + 1.25)⁵]. The government, not the firm, realizes this social impact value. Effectively, the government has lost $100 in foregone revenue — borrowed at a 5% annual interest rate to make up for that foregone revenue — and received roughly $305 in value. This is an extremely attractive proposition at scale, even with artificially high costs for government lending (5% is too high, but it’s a good illustrative number).
The problematic parts of the proposal would be: (1) the required government lending to fund the program upfront; and (2) measuring the aggregate returns of the program. (Both of these are solvable problems that I’ll save for a future essay). The tricky aspect of this program would be tracking the various flows of social impact value.
For example, imagine that the impact tax incentive program has become ubiquitous and funds social impact generation across various parts of life: schools, prisons, hospitals, etc. To be truly effective, the program would need to attribute each bit of value created back to relevant programs or agencies so as to appropriately reduce budgets. To give a concrete example, a a nationally scaled program would need to be able to determine how much value came explicitly from a reduction in recidivism rates of inmates currently housed in the federal prison system (or any other unit of specificity, perhaps in education or healthcare). This specificity would allow the budgets for each recipient of federal funding (e.g., the Department of Justice) to decrease by the amount equivalent to the social value generated by the relevant impact-generating firm. Any offsetting increase would be independent of the firm that received the impact-generator tax incentive.
At best, this program could represent massive savings and overwhelmingly encourage investors to pursue impact initiatives because of the growth advantage from preferential tax treatment. While not receiving a dividend outright, investors would receive adequate compensation in the form of growth ability. The benefits to the governments currently paying for budgets that would ultimately be reduced because of the effects of impact-generating companies would be at worst, marginal, and at best, a windfall.
