Subsidy and Information to Increase Therapy Demand
Introduction
Roth et. al. (2024) executes a randomized controlled trial to estimate the effect of information on depressed individuals’ willingness to pay for therapy.
The RCT assigns a sample of depressed people who have never tried therapy before to either a control group or an information group. Only the information group is told that a review of the literature shows that “out of the 22 clinical studies with at least a hundred participants, all 22 studies find that therapy is an effective treatment for depression.” Both groups then complete survey questions that elicit their willingness-to-pay for therapy using the Becker-DeGroot-Marschak mechanism. The difference between these two groups’ willingness-to-pay is $10 out of a $320 cost for four weeks of therapy. The paper interprets this as evidence that information increases demand for therapy by shifting people’s attention and beliefs.
The survey allows the paper to estimate the MVPF for a combined policy providing both information about the effectiveness of therapy for treating depression and subsidies for therapy. The paper estimates MVPFs under a variety of subsidy levels; in the description below, we focus on a $70 subsidy for the $320 cost of four weeks of therapy.
The paper also constructs separate MVPFs for the subsidy policy and information policy.
MVPF = 2.1
Net Cost
The paper estimates the direct cost of the subsidy and information policy as the rate of subsidy take-up multiplied by the subsidy amount, as well as the cost of providing information.
The paper assumes that the price of four weeks of online therapy is $320. At a $70 subsidy level, 21% of people in the information group have a willingness-to-pay above the cost of therapy and would therefore take-up the subsidy. Thus, the direct cost for the subsidy per individual treated with information is $70*0.21 = 14.7.
Assuming the direct cost of providing information is $1, the combined direct cost for the subsidy and information policy is $15.70.
In some of the estimates, the paper additionally includes a fiscal externality of increased tax revenues due to productivity gains from depressed individuals seeking therapy. Assuming an effective 20% tax rate on the gains accruing due to marginal individuals yields increased tax revenue of $11.70. This assumption yields a net cost of $4.
Finally, the paper accounts for a scenario where depressed individual seeking therapy will have a positive externalities on other individuals such as friends or employers. Including the assumption that these positive externalities would yield additional productivity gains leads to an additional $5.86 in tax revenue and a negative net cost.
Willingness to Pay
To estimate the benefits for this policy accruing to private individuals the paper considers two groups of people: those who would take-up therapy without information or a subsidy (inframarginals), and those who are induced to take-up therapy by the combination of information and a subsidy (marginals). The paper defines a marginal individual as someone whose individual willingness to pay (WTP) is above the subsidized cost but below the raw cost. The survey elicitation allows the paper to estimate that 16% of individuals are marginal and 5% of individuals are inframarginal to the combined policy of information and a $70 subsidy.
Marginal individuals gain the surplus of their individual willingness-to-pay minus the subsidized cost of therapy. The paper also assumes these marginal individuals seeking therapy will have a positive externality on other individuals such as friends or employers. The paper estimates this externality as an additional benefit equivalent to 50% of the individual’s willingness to pay. The average individual willingness-to-pay for a marginal individual is $292.6. This means that the total benefits of the policy due to a marginal individual are (1+0.5)*292.6- ($320 – $70) = $188.90.
The inframarginals seek out therapy in the absence of the subsidy and information policy, so the benefits due to an inframarginal individual is solely the $70 transfer from the subsidy.
The paper then weights these two values by the share of individuals who are marginal and inframarginal leading to a willingness-to-pay of:
(0.16*$188.90) + (0.05*$70) = $33.70
In some of estimates, the paper conservatively estimates there are no positive externalities. That assumption leads to a willingness-to-pay of $10.32.
MVPF Summary
The MVPF for the combined policy of information and a $70 subsidy under the assumptions of no productivity impacts and a 50% externality is 2.140.
- Assuming productivity impacts and a 50% externality leads to an infinite MVPF.
- Assuming productivity impacts but not externality leads to an MVPF estimate of 2.543
- Assuming no productivity impacts and no externalities that leads to an MVPF of 0.658.
The paper offers MVPF estimates varying other assumptions as well:
- Estimating the direct cost of the information as $5 per person (instead of $1) leads to MVPFs ranging from 0.523 to 14.8.
- Varying the subsidy level from $20 to the full cost of $320 (holding the assumptions of no productivity impacts and a 50% externality fixed), the MVPF estimates are decreasing in the level of the subsidy from 3.158 to 0.819.
References
Roth, Christopher, Peter Schwardmann, and Egon Tripodi (2024). “Misperceived Effectiveness and the Demand for Psychotherapy.” Journal of Public Economics: Vol. 240 (105254). DOI: https://doi.org/10.1016/j.jpubeco.2024.105254
Policy Features
- Category
- In Kind Transfers
- Sub-Category
- Health Adult
- Beneficiary Type(s)
- Adults
- Average Age
- 30
- Country of Implementation
- United States
- Empirical Method
- RCT
- Research Type
- Primary
- Peer Reviewed
- No
- MVPF Publication Link
- doi.org/10.1016/j.jpubeco.2024.105254