Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Sign Up
PreprintCited by 5

The DOI matches its Crossref record.

Discrimination-Free Insurance Pricing

Lindholm, Mathias, Richman, Ronald, Tsanakas, Andreas, Wuthrich, Mario V.·2020
DOI 10.2139/ssrn.3520676
Open this paper in CoChat — free account

Abstract

A simple formula for non-discriminatory insurance pricing is introduced. This formula is based on the assumption that certain individual (discriminatory) policyholder information is not allowed to be used for insurance pricing. The suggested procedure can be summarized as follows: First, we construct a price that is based on all available information, including discriminatory information. Thereafter, we average out the effect of discriminatory information. This averaging out is done such that discriminatory information can also not be inferred from the remaining non-discriminatory one, thus, neither allowing for direct nor for indirect discrimination.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.2139/ssrn.3520676

Cites

Statistics

Figures

FIGURE 1 (left) Causal diagram described by G ; (right) causal diagram altered according to the intervention X = x .
FIGURE 1 (left) Causal diagram described by G ; (right) causal diagram altered according to the intervention X = x .CC-BY · From the PDF
FIGURE 2 True model: (left) smokers and (right) non-smokers with solid black and red lines giving the best-estimate prices for women and men, respectively. The dotted orange lines show the discrimination-free prices and the dotted blue lines show the unawareness prices.
FIGURE 2 True model: (left) smokers and (right) non-smokers with solid black and red lines giving the best-estimate prices for women and men, respectively. The dotted orange lines show the discrimination-free prices and the dotted blue lines show the unawareness prices.CC-BY · From the PDF
FIGURE 3 True model: (left) smokers and (right) non-smokers with solid black and red lines giving the best-estimate prices for women and men, respectively. The dotted orange lines show the discrimination-free prices and the dotted green lines show the unawareness prices, for an alternative assumption on P ( D = woman | smoker).
FIGURE 3 True model: (left) smokers and (right) non-smokers with solid black and red lines giving the best-estimate prices for women and men, respectively. The dotted orange lines show the discrimination-free prices and the dotted green lines show the unawareness prices, for an alternative assumption on P ( D = woman | smoker).CC-BY · From the PDF
FIGURE 4 The age frequency used for both genders and smoking habits to simulate the data.
FIGURE 4 The age frequency used for both genders and smoking habits to simulate the data.CC-BY · From the PDF
FIGURE 5 Estimated regression functions ̂ λ 1 ( X , D ) (left), ̂ λ 2 ( X , D ) (middle), and ̂ λ 3 ( X , D ) (right) using the neural network architecture of Listing 2.
FIGURE 5 Estimated regression functions ̂ λ 1 ( X , D ) (left), ̂ λ 2 ( X , D ) (middle), and ̂ λ 3 ( X , D ) (right) using the neural network architecture of Listing 2.CC-BY · From the PDF
FIGURE 6 Estimated neural network model: (left) smokers and (right) non-smokers with solid black and red lines giving the best-estimate prices for women and men, respectively. The dotted orange lines show the discrimination-free prices and the dotted blue lines show the unawareness prices.
FIGURE 6 Estimated neural network model: (left) smokers and (right) non-smokers with solid black and red lines giving the best-estimate prices for women and men, respectively. The dotted orange lines show the discrimination-free prices and the dotted blue lines show the unawareness prices.CC-BY · From the PDF
FIGURE 7 Estimated regression functions ̂ λ 1 ( X ) (left), ̂ λ 2 ( X ) (middle), and ̂ λ 3 ( X ) (right) using neural networks and ignoring the gender information D .
FIGURE 7 Estimated regression functions ̂ λ 1 ( X ) (left), ̂ λ 2 ( X ) (middle), and ̂ λ 3 ( X ) (right) using neural networks and ignoring the gender information D .CC-BY · From the PDF
FIGURE 8 GLMestimated regression functions ̂ λ GLM 1 ( X , D ) (left), ̂ λ GLM 2 ( X , D ) (middle) and ̂ λ GLM 3 ( X , D ) (right).
FIGURE 8 GLMestimated regression functions ̂ λ GLM 1 ( X , D ) (left), ̂ λ GLM 2 ( X , D ) (middle) and ̂ λ GLM 3 ( X , D ) (right).CC-BY · From the PDF
FIGURE 9 Estimated GLM: (left) smokers and (right) non-smokers with solid black and red lines giving the best-estimate prices for women and men, respectively. The dotted orange lines show the discrimination-free prices and the dotted blue lines show the unawareness prices.
FIGURE 9 Estimated GLM: (left) smokers and (right) non-smokers with solid black and red lines giving the best-estimate prices for women and men, respectively. The dotted orange lines show the discrimination-free prices and the dotted blue lines show the unawareness prices.CC-BY · From the PDF
FIGURE 10 Bias-corrected discrimination-free prices ̂ h ∗ ( x ) against unadjusted discrimination-free prices h ( x ).
FIGURE 10 Bias-corrected discrimination-free prices ̂ h ∗ ( x ) against unadjusted discrimination-free prices h ( x ).CC-BY · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

No references could be resolved yet.

Research works printed but not linked: 22

  1. ASEERVATHAM, V., LEX, C. and SPINDLER, M. (2016) How do unisex rating regulations affect gender differences in insurance premiums? The Geneva Papers on Risk and Insurance-Issues and Practice 41(1), 128-160.
  2. AVRAHAM, R., LOGUE, K. D. and SCHWARCZ, D. B. (2014) Understanding insurance anti- discrimination laws. Southern California Law Review 87(2), 195-274.
  3. BREUER, T. and CSISZÁR, I. (2013) Systematic stress tests with entropic plausibility constraints. Journal of Banking & Finance 37(5), 1552-1559.
  4. CHEN, A., GUILLÉN, M. and VIGNA, E. (2018) Solvency requirement in a unisex mortality model. ASTIN Bulletin: The Journal of the IAA 48(3), 1219-1243.
  5. CSISZÁR, I. (1975) I-divergence geometry of probability distributions and minimization prob- lems. The Annals of Probability 3(1), 146-158.
  6. DE JONG, P. and FERRIS, S. (2006) Adverse selection spirals. ASTIN Bulletin: The Journal of the IAA 36(2), 589-628.
  7. European Commission (2012). Guidelines on the application of COUNCIL DIRECTIVE 2004/113/EC to insurance, in the light of the judgment of the Court of Justice of the European Union in Case C-236/09 (Test-Achats). Official Journal of the European Union C11, 1-11.
  8. European Council (2004) COUNCIL DIRECTIVE 2004/113/EC -implementing the principle of equal treatment between men and women in the access to and supply of goods and services. Official Journal of the European Union L 373, 37-43.
  9. FREES, E. W. and HUANG, F. (2021) The discriminating (pricing) actuary. North American Actuarial Journal, to appear.
  10. FRIEDMAN, J. H. (2001) Greedy function approximation: a gradient boosting machine. Annals of Statistics, 1189-1232.
  11. FUSTER, A., GOLDSMITH-PINKHAM, P., RAMADORAI, T. and WALTHER, A. (2018). Predictably unequal? the effects of machine learning on credit markets. Available at SSRN: https://ssrn.com/abstract=3072038 (Downloaded on Feb. 21, 2020).
  12. GUELMAN, L. and GUILLÉN, M. (2014). A causal inference approach to measure price elasticity in automobile insurance. Expert Systems with Applications 41(2), 387-396.
  13. GUILLÉN, M. (2012). Sexless and beautiful data: from quantity to quality. Annals of Actuarial Science 6(2), 231-234.
  14. KRIKLER, S., DOLBERGER, D. and ECKEL, J. (2004) Method and tools for insurance price and revenue optimisation. Journal of Financial Services Marketing 9(1), 68-79.
  15. KUSNER, M. J., LOFTUS, J., RUSSELL, C. and SILVA, R. (2017) Counterfactual fairness. In Advances in Neural Information Processing Systems, pp. 4066-4076.
  16. LAURITZEN, S. L. (1996) Graphical Models. Oxford Science Publications.
  17. PEARL, J., GLYMOUR, M. and JEWELL, N. P. (2016) Causal Inference in Statistics: A Primer. John Wiley & Sons.
  18. POPE, D. G. and SYDNOR, J. R. (2011) Implementing anti-discrimination policies in statistical profiling models. American Economic Journal: Economic Policy 3(3), 206-231.
  19. PRINCE, A. E. R. and SCHWARCZ, D. (2019) Proxy discrimination in the age of artificial intelligence and big data. Iowa Law Review 105, 1257.
  20. SASS, J. and SEIFRIED, F. T. (2014) Insurance markets and unisex tariffs: Is the European Court of Justice improving or destroying welfare? Scandinavian Actuarial Journal 2014(3), 228-254.
  21. WILLIAMS, D. (1991) Probability with Martingales. Cambridge University Press.
  22. ZHAO, Q. and HASTIE, T. (2021) Causal interpretations of black-box models. Journal of Business & Economic Statistics 39(1), 272-281.

Cited by

Metadata

Authors
Lindholm, M., Richman, R., Tsanakas, A., & Wuthrich, M. V.
ORCID
Tsanakas, Andreas
Publisher
Elsevier BV
Published
2020
DOI
10.2139/ssrn.3520676
Cited by
5
References
0
Type
Preprint

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.