Discounting health and money: New evidence using a more robust method

This study compares discounting for money and health in a field study. We applied the direct method, which measures discounting independent of utility, in a representative French sample, interviewed at home by professional interviewers. We found more discounting for money than for health. The median discount rates (6.5% for money and 2.2% for health) were close to market interest rates, suggesting that at the aggregate level the direct method solves the puzzle of unrealistically high discount rates typically observed in applied economics. Constant discounting fitted the data better than the hyperbolic discounting models that we considered. The substantial individual heterogeneity in discounting was correlated with age and occupation. Electronic supplementary material The online version of this article (10.1007/s11166-018-9279-1) contains supplementary material, which is available to authorized users.

A Example of choice questions used in the experiment  Figure 1 in the main text.

G1 -Which do you prefer?
Option A A 20% increase in your purchasing power between age 40 and 50 Option B A 20% increase in your purchasing power between age 50 and 60

I choose A I choose B
To help you make your choice, the options are displayed visually in the graph below

B Rounding, iterations and indifferences
In this Section, we use the data from a selected participant to illustrate how we dealt with rounding. We then provide some descriptive statistics.
B.1 Data from a selected participant Table B.1 shows the choices faced by subject ID1802401 and her/his answers in the elicitation of t .5 , for monetary outcomes. The value determined by three indifferences was equal to 42. In the analysis reported in the main text, the value was determined after four iterations and was equal to 43.

C.2 Constant discounting
The best-known case of d t is constant discounting: time weights are equal to e −δt , where δ is the discount rate. Therefore, using a continuous approximation: with c a normalization constant. The primitive of e −δt is − 1 δ e −δt . Then: This gives C(0) = 0 and C(T ) = c δ 1 − e −δT . Defining the normalization constant such that C(T ) = 1: one gets: (2)

D Empirical Cumulative Distribution Functions for Cumulative Weights
• What is your age?
• What is your main occupation?
• Are you single or living as a couple?
• If you are living as a couple, what is your partner's main occupation?
We will now ask you a series of questions based on different scenarios. One scenario will be about health and quality of life. One scenario will be about your purchasing power. One scenario will be about your diet.
Imagine you suffer from back pain. Back pain corresponds to a mild but continuous pain. In practice this corresponds to: • some problems walking about: unable to remain a long time in sitting or standing position, walking for a long time, unable to carry heavy loads.
• no problems with performing self care activities (e.g. eating, washing or dressing).
• some problems with performing usual activities (e.g. work, study, housework, family or leisure activities).
• mild pain or discomfort.
• not anxious or depressed.
Imagine that you suffer from back pain and that there is a treatment that allows complete relief. This treatment involves a weekly dose of pills, which would result in full health. Without the treatment, you will suffer again from the same above-mentioned back pain. Imagine that you have the possibility to take this treatment, but only for a given period of time.
We will use the computer to answer a series of questions. The health scenario is described on a cardboard card to which you can refer as often as you need during the series of questions.
Questions are about your preference between two periods of time when to take the treatment for complete back pain relief. Imagine that you can be relieved from back pain between your current age and your age+10 years, or between your current age+10 years and your current age +20 years. Of these two periods, in which one do you prefer to be released from back pain? Note that beyond this period you will still suffer from back pain. You will face a series of successive questions about your preference between two periods, with different lengths for the periods. Take your time to answer each of the questions. There are no right or wrong answers.
To help you express your preference, a strip placed at the bottom of the screen summarizes these two options. The light gray part corresponds to option A, whereas the dark gray portion corresponds to option B. If you choose option A (option B), it means that you will suffer from back pain during the time period in dark gray (light gray).
At the end of each series of questions, the first question will be repeated. If you answer differently to this repeated question, we will ask you to express your preferences again.
We will now ask you another series of questions relating this time to a scenario about purchasing power. Your purchasing power indicates what your income allows you to buy. Imagine your purchasing power is, for a given period of time, 20% higher compared to your purchasing power today and that without changing anything in your current activities (e.g without further work). This 20% increase in your purchasing power only occurs for a period of time. At the end of that period your purchasing power will return to its normal level. Now we will use the computer to answer a series of questions. The purchasing power scenario is described on the following cardboard card to which you can refer as often as you need during the series of questions.
Questions are about your preference between two periods. Imagine your purchasing power can increase by 20% between your age and your age+10 years, or between your current age+10 years and your current age+20 years. Of these two periods, in which one do you prefer to have your purchasing power 20% higher compared to its current level? Note that beyond this period your purchasing power will be at the same level as it is now. You will face a series of successive questions about your preference between two periods, with different lengths for the periods. Take your time to answer to each of the questions. There are no right or wrong answers.
To help you express your preference, a strip placed at the bottom of the screen summarizes these two options. The light gray part corresponds to option A, whereas the dark gray portion corresponds to option B. If you choose option A (option B), it means that you have a 20% increase in your purchasing power during the time period in light gray (dark gray).
At the end of each series of question, the first questions will be repeated. If you answer differently to this repeated question, we will ask you to express your preferences again.

E.2 Additional questions and questionnaire
Subjects also faced a third scenario. This scenario was similar to the abovementioned scenarios except that it was based on a choice between the two periods in which the subject was supposed to follow a low-salt diet allowing him to stay healthy in the future.
In addition to the scenarios, participants were asked a series of questions about their health status, their perception of the future, their tobacco and alcohol consumption, their perceptions of the risks associated with tobacco and alcohol consumption and their diet. The last series of questions was about their educational degree (if any), their financial situation and whether they sufferred from back pain or followed a low-salt diet.