New WTP Estimates of the Demand for Food Safety

James K. Hammitt

Kevin Haninger

Harvard Center for Risk Analysis

2

Outline

Motivation

Survey instrument and sample

Descriptive statistics

WTP estimates

Conclusion

3

Motivation

Most cases of food-borne illness are non-fatal, short duration, caused by microbial pathogens

There are few estimates of WTP to reduce risk of comparatively minor morbidity

We conducted stated preference (contingent valuation) study of WTP to reduce risk of food-borne illness, focusing on acute episodes

4

Survey Administration

Knowledge Networks internet panel• Representative of general population (on measurable

attributes)• 61% of invited panel members completed survey• 3% excluded (did not eat any of three foods or did not

answer WTP questions)Intervention: choose food produced by “superior

safety system” (safer, more expensive) or conventional

Double-bounded dichotomous-choice valuation questions

5

Survey Instrument

Each respondent values two changes• Own risk

• Risk to child (2 – 18 yrs) in household (if applicable)

• Choice described as “food only [you / your child] will eat”

Risk of illness from microbial pathogens on food

Risk reduction and cost expressed per meal or per month (28% of respondents)

6

Risk Attributes

Risk reduction: [4 or 2] to 1 per 10,000 per meal• Risk reduction = [3 or 1] per 10,000• Visual aid: white square w/ fraction colored red

Duration: 1, 3, 7 days

Severity: mild, moderate, severe

Mortality risk conditional on illness: 0, 1/10,000, 1/1,000

Food: chicken, ground beef, packaged deli meat

Initial bid: between $0.04 and $4.00 per meal• Follow-up bids half and twice as large

Per month: risk and bid calculated using respondent-reported consumption frequency

7

Severity1. You will have an upset stomach and will feel tired, but these

symptoms will not prevent you from going to work or from doing most of your regular activities.

2. You will have an upset stomach, fever, and will need to lie down most of the time. You will be tired and will not feel like eating or drinking much. Occasionally, you will have painful cramps in your stomach. In addition, you will have some diarrhea and will need to stay close to a bathroom. While you are sick, you will not be able to go to work or do most of your regular activities.

3. You will have to be admitted to a hospital. You will have painful cramps in your stomach, fever, and will need to spend most of your time lying in bed. You will need to vomit and will have severe diarrhea that will leave you seriously dehydrated. Because you will be unable to eat or drink much, you will need to have intravenous tubes put in your arm to provide nourishment.

8

Structure of Survey

Introduction• Experience with food-borne illness• Estimated prevalence

Practice valuation questions• Similar to elicitation• 1st practice: dominant choice, feedback to respondent, opportunity

to correct dominated choice• 2d practice: no dominance but feedback to respondent interpreting

choice (i.e., “safer but more expensive,” “less safe but less expensive”)

Valuation questionsFollow-up questions

• Scenario acceptance• Food-handling practices

9

Type of chicken Your chance of illness per meal Cost per meal

Standard 4 in 10,000 Standard cost

Superior safety system 1 in 10,000 $0.50 more per meal than standard

The table below summarizes the differences between the Superior Safety System chicken and the standard chicken. Please consider which type of chicken you would buy for a meal that only you would eat. Remember that the extra money you spend for a meal with Superior Safety System chicken is money that you could no longer spend on other things you might want or need.

Whether you eat the Superior Safety System chicken or the standard chicken, if you get sick:You will have an upset stomach, fever, and will need to lie down most of the time. You will be tired and will not feel like eating or drinking much. Occasionally, you will have painful cramps in your stomach. In addition, you will have some diarrhea and will need to stay close to a bathroom. While you are sick, you will not be able to go to work or do most of your regular activities.You will have these symptoms for 3 days.There is a 1 in 10,000 chance that you will die from this sickness.If Superior Safety System chicken cost $0.50 more per meal than standard chicken, which type of chicken would you purchase?○ Standard○ Superior Safety System

10

Sample Mean (std dev)Demographics Perception, behavior

Age 45 (16) Perceived prevalence 0.33 (0.23)

Child’s age 8.5 (4.9) Own risk higher 0.13

Male 0.48 Own risk lower 0.39

Black, non-Hispanic 0.11 Child’s risk higher 0.15

Hispanic 0.12 Child’s risk lower 0.34

Married 0.54 Prior illness 0.38

Household size 2.6 (1.4) Not confident in safety 0.16

Log HH income 10.4 (0.9) Distrust government 0.04

College degree 0.25 Distrust private sector 0.06

Consumption/month 5.7 (6.3) Prepare HH food (0-4) 2.36 (1.47)

Wash hands 0.62

11

Household Comparison(means)

With child Without

Age 37 48

Married 0.68 0.48

Household size 3.6 2.2

College degree 0.21 0.27

12

Estimating WTP

Double-bounded binary-choice response

Maximum-likelihood regression estimates assuming lognormal error

Separate models and subsamples• Reduce own risk, no child in household (N = 2467)

• Reduce own risk, child in household (N = 1160)

• Reduce risk to child (between 2 and 18)

13

Regression Estimates

Adult w/o child

(N = 4934)

Adult w/ child

(N = 1160)

Child

(N = 1149)

Intercept 0.06 (0.11) 0.62 (0.25)*** 1.19 (0.25)***

Risk redn = 3 0.62 (0.08)*** 0.20 (0.16) 0.45 (0.17)***

Moderate 0.30 (0.10)*** 0.21 (0.20) 0.04 (0.21)

Severe 0.39 (0.10)*** 0.39 (0.20)*** 0.03 (0.20)

3 days 0.06 (0.10) -0.06 (0.20) 0.08 (0.20)

7 days 0.25 (0.10)*** -0.06 (0.20) -0.11 (0.20)

Mort 1e-4 0.12 (0.10) -0.05 (0.20) -0.09 (0.20)

Mort 1e-3 0.03 (0.10) 0.01 (0.20) -0.27 (0.20)

Per month 1.77 (0.09)*** 1.68 (0.18)*** 1.63 (0.18)***

14

Sensitivity of WTP

More sensitive to probability change than to severity or duration of illness• Less than proportional to risk change (for adult

w/o child, WTP varies by 1.9, risk by 3)

Less sensitive to severity and duration for children than for adults

WTP per month proportional to frequency of consumption

15

Demographics & Food Type

Adult w/o child Adult w/ child Child

Age 0.010 (0.003)*** 0.007 (0.008) 0.017 (0.019)**

Age of child -0.032 (0.018)*

Male -0.42 (0.08)*** -0.25 (0.18) -0.54 (0.19)***

Male child -0.01 (0.16)

Black, non-H 0.86 (0.14)*** 1.21 (0.29)*** 0.83 (0.29)***

Hispanic 0.62 (0.13)*** 0.72 (0.24)*** 0.55 (0.24)**

Log HH income 0.03 (0.05) 0.04 (0.10) -0.15 (0.10)

College degree -0.36 (0.09)*** -0.48 (0.21)** -0.46 (0.21)**

Ground beef -0.38 (0.09)*** -0.46 (0.19)** 0.26 (0.20)

Deli meat -0.32 (0.13)** -0.58 (0.29)** -0.35 (0.29)

16

Risk Perception & Behavior

Adult w/o child Adult w/ child Child

Perceived prevalence 0.013 (0.002)*** 0.001 (0.004) 0.002 (0.004)

Perceive higher risk 0.39 (0.13)*** 0.65 (0.26)*** 0.34 (0.25)

Perceive lower risk -0.28 (0.09)*** -0.26 (0.17) -0.65 (0.18)***

Prior illness -0.21 (0.08)** 0.02 (0.16) -0.02 (0.17)

Not confident of safety -1.04 (0.12)*** -0.63 (0.24)*** -0.52 (0.24)**

Distrust government -0.71 (0.20)*** 0.12 (0.45) -0.18 (0.46)

Distrust private sector 0.49 (0.18)*** -0.07 (0.39) -0.21 (0.39)

Prepare HH food -0.10 (0.03)*** -0.01 (0.07) 0.02 (0.07)

Wash hands 0.42 (0.08)*** 0.53 (0.16)*** 0.08 (0.17)

Safe practices 0.22 (0.12)* 0.08 (0.26) -0.04 (0.27)

2d risk question 0.19 (0.08)*** 0.58 (0.16)*** -0.17 (0.16)

17

Value per Statistical Case

Median WTP (lognormal error)

Sample-mean respondent

Specified illness

Mean of value per case for larger and smaller risk reductions

18

Value per Statistical Case($1,000, std. err.)

Severity DurationAdult w/o children

Adult w/ children Child

Mild 1 8.3 (0.7) 10.8 (0.9) 24.9 (7.0)

Moderate 1 11.1 (0.9) 13.1 (1.1) 28.0 (7.7)

Severe 1 12.5 (1.0) 16.4 (1.3) 28.1 (7.6)

Mild 3 8.8 (0.7) 10.2 (0.9) 27.0 (7.4)

Moderate 3 11.7 (1.0) 12.3 (1.1) 30.4 (8.3)

Severe 3 13.2 (1.1) 15.4 (1.3) 30.5 (8.3)

Mild 7 10.8 (0.9) 10.4 (1.0) 23.6 (6.6)

Moderate 7 14.4 (1.1) 12.5 (1.3) 26.5 (7.5)

Severe 7 16.1 (1.2) 15.7 (1.5) 26.7 (7.3)

19

Value per Statistical Life

WTP not significantly related to conditional mortality risk

Point estimates: $9 – 25 million • Study underpowered for estimating VSL

20

WTP per QALY

Estimated value per case varies little with duration (1 – 7 days)• 28% more for 7 days than 1 day (adult w/o child)

Varies little with severity• 48% more for severe than mild case (adult w/ or w/o child)

Elicited health-related quality of life (HRQL) for current health and each severity• Loss in HRQL 1.9 to 2.3 times larger for severe than mild case

Value per case varies much less than proportionately to loss of HRQL

21

HRQL by Severity(mean, std dev)

Health Utilitities

Index DecrementVisual

analog scale Decrement

Current health

0.80 (0.21) 0.76 (0.17)

Mild 0.51 (0.27) 0.29 0.58 (0.21) 0.18

Moderate 0.26 (0.31) 0.54 0.47 (0.23) 0.29

Severe 0.12 (0.31) 0.68 0.42 (0.25) 0.34

22

Conclusion: WTP per case of acute food-borne illness

Varies with potential victim• $8,000 – 16,000 for adult respondent• $24,000 – 31,000 for his/her child

Varies with food type, respondent characteristics• Higher for chicken than ground beef, packaged deli meat• Higher for women, blacks and Hispanics, less educated• Increases with perceived risk and confidence in intervention

Largely insensitive to duration & severity• ≤ 30% larger for a week than a day• ≤ 50% larger for severe than mild• Less sensitive for risk to child than to adult

Estimated WTP is less than proportional to expected loss in QALYs