Skip to main content

Advertisement

SpringerLink
Log in

Insurance Search, Switching Behavior and the Role of Group Contracts

  • Published:
De Economist Aims and scope Submit manuscript

Abstract

This paper considers consumer behavior in the Dutch market for health insurances, which is characterized by managed competition. We observe substantial price dispersion for the same basic insurance package. We describe the market by a simple consumer search model, which we empirically test using data from the Dutch Health Care Consumer Panel. We argue that insurers use group contracts to target discounts to better informed consumers, reducing their incentives to search. Since search is essential for competition, this increases premiums and premium dispersion.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Similar content being viewed by others

Notes

  1. Some civil servants (for example, the police force) were covered by a compulsory insurance scheme irrespective of their income. This concerned about 5 % of the total population.

  2. The contribution was 7.95 % of income, of which 6.25 % was paid by the employer.

  3. Chronically ill with a high income, who would be refused by private insurers were covered by a special insurance.

  4. Most employees received compensation of about 50 % of the premium from their employer.

  5. For risk equalization the Risk Equalization Fund used in 2006 gender-age interactions, long-term medication use (categorized in 17 groups), medical diagnosis (clustered in 13 groups), region (clustered in 10, not necessarily geographical, clusters) and main source of income interacted with age. Data from 2003 were used to calculate the compensation levels. A more detailed description of the Risk Equalization Fund can be found in “Appendix 2”.

  6. Children under age 18 are covered by their parents’ insurance and their premium is paid by the government.

  7. There is no systematic difference in price level between former Sickness Funds and private insurers.

  8. The information on the price of all available insurance packages in the market used for the figures has been supplied by Zorgweb.

  9. Most insurers already announced the premium for the basic insurance in October or November 2005. However, some insurers lowered their premium after learning the premiums of their competitors.

  10. It was also announced that in later years insurers could deny supplemental insurance coverage for new clients.

  11. As required by law, such targeted insurance packages were open to everyone. However, for this particular insurance targeted towards students, 75 % of new enrollees were between the age of 18 and 25.

    Fig. 6
    figure 6

    Boxplot of the price for basic insurance without discounts

    Full size image

  12. The assumption that switching is costless can be relaxed, maintaining qualitatively similar results.

  13. The Dutch government launched this website to stimulate competition by increasing transparency.

  14. If an individual also received an offer for a group contract, there are in fact only \(N-2\) other insurers. Only if the offer for a group contract is with the same insurer the individual was previously insured with, there are still \(N-1\) other insurers. For ease of exposition we ignore this, as taking account of this complicates notation without changing our testable predictions.

  15. The model could allow for switching costs as well. In that case the individual only switches to the insurer with the lowest premium if that premium is sufficiently below the current best offer.

  16. Obviously, marginal costs should depend on the health status of the insuree. However, recall that the Risk Equalization Fund compensates insurers for insuring individuals in bad health in such way that the expected costs of all insurees are the same.

  17. For later years this information is available, but because of multiple mergers between health insurers not insightful for 2006.

  18. We tried weighting observations, but this did not affect any of our results.

  19. We do not observe income directly, but rather observe the amount of government compensation an individual receives. Very low income households (less than €17,500 per year) receive the maximum annual compensation of €402.96 for a single, and €1,155.00 for a couple. Partial (income dependent) compensation was paid to low income household (below €25,068 for singles and €40,120 for couples).

    Table 3 Descriptive statistics
    Full size table

  20. These are national level figures. In our sample we observe that 85 % of the group contracts are obtained via the employer.

  21. Since individuals are replaced every two to three years, this information is at most three years old.

  22. The Pharmaceutical Cost Groups use information on medication that was prescribed for at least 181 days in the past year as an indicator for chronic diseases. Individuals with different types of medication (and all used for at least 181 days) are classified in the PCG of the most expensive drug they use.

  23. The Diagnosis Cost Groups use the diagnosis from hospital admissions to create clusters of diagnoses based on the costs involved. Again, if an individual qualifies for more than one DCG, she is classified in the highest cost DCG.

References

  • Brown, J. R., & Goolsbee, A. (2002). Does the internet make markets more competitive? Evidence from the life insurance industry. Journal of Political Economy, 110, 481–507.

    Article  Google Scholar 

  • Cameron, A. C., Trivedi, P. K., Milne, F., & Piggott, J. (1988). A microeconometric model of the demand for health care and health insurance in Australia. Review of Economic Studies, 55, 85–106.

  • Carlson, J. A., & McAfee, R. P. (1983). Discrete equilibrium price dispersion. The Journal of Political Economy, 91, 480–493.

    Article  Google Scholar 

  • Conniffe, D., & O’Neill, D. (2008). An efficient estimator for dealing with missing data on explanatory variables in a probit choice model, NUIM Working Paper #N1960908.

  • Dutch Health Care Authority (NZa, Nederlandse Zorgautoriteit). (2006). De tussenstand op de verzekeringsmarkt. Monitor zorgverzekeringsmarkt juni 2006 (The interim score on the health insurance market. Monitor health insurance market June 2006).

  • Dutch Health Care Authority (NZA, Nederlandse Zorgautoriteit). (2007). Richting geven aan keuzes (Directing choices).

  • Dutch Health Care Authority (NZA, Nederlandse Zorgautoriteit). (2009). Monitor zorgverzekeringsmarkt. Antwoorden op vragen van het ministerie van VWS over toetreding, verticale integratie en koppelingen aanvullende verzekering basisverzekering (monitor health Insurance market. Answers to questions of the Ministry of Health, Welfare and Sport about entry, vertical integration and bundling supplementary insurance and basic insurance).

  • Dutch Health Care Authority (NZA, Nederlandse Zorgautoriteit). (2011). Marktscan Zorgverzekeringsmarkt. Weergave van de markt 2007–2011 (market scan health insurance market. Representation of the market 2007–2011).

  • Etnner, S. L. (1997). Adverse selection and the purchase of medigap insurance by the elderly. Journal of Health Economics, 16, 543–562.

    Article  Google Scholar 

  • Gaynor, M., Moreno-Serra, R., & Propper, C. (2010). Death by market power: reform, competition and patient outcomes in the national health service, NBER working paper 16164.

  • Janssen, M. C. W., & Moraga-González, J. L. (2004). Strategic pricing, consumer search and the number of firms. Review of Economics and Statistics, 71, 1089–1118.

    Article  Google Scholar 

  • Janssen, M. C. W., Moraga-González, J. L., & Wildenbeest, M. R. (2005). Truly costly sequential search and oligopolistic pricing. International Journal of Industrial Organisation, 23, 451–466.

    Article  Google Scholar 

  • Maestas, N., Schroeder, M., & Goldman, D. (2009). Price variation in markets with homogeneous goods: The case of medigap, NBER working paper 14679.

  • Ministry of Health, Welfare and Sport. (2008). Risk adjustment under the Health Insurance Act in the Netherlands.

  • Pauly, M. V., Herring, B., & Song, D. (2006). Information technology and consumer search for health insurance. International Journal of the Economics of Business, 13, 45–63.

    Article  Google Scholar 

  • Salop, S., & Stiglitz, J. (1977). Bargains and ripoffs: A model of monopolistically competitive price dispersion. Review of Economic Studies, 44, 493–510.

    Article  Google Scholar 

  • Savage, E., & Wright, D. J. (2003). Moral hazard and adverse selection in australian private hospitals: 1989–1990. Journal of Health Economics, 22, 331–359.

    Article  Google Scholar 

  • Sorensen, A. T. (2000). Equilibrium price dispersion in retail markets for prescription drugs. Journal of Political Economy, 108, 833–850.

    Article  Google Scholar 

  • Sorensen, A. T. (2001). An Empirical model of heterogeneous consumer search for retail prescription drugs, NBER working paper 8548.

  • Stahl, D. O. (1989). Oligopolistic pricing with sequential consumer search. American Economic Review, 79, 700–712.

    Google Scholar 

  • Stahl, D. O. (1996). Oligopolistic pricing with heterogeneous consumer search. International Journal of Industrial Organization, 14, 243–268.

    Article  Google Scholar 

  • Van Kleef, R. C., van Vliet, R. C. J. A., & van de Ven, W. P. M. M. (2012). Risicoverevening 2012. Een analyse van voorspelbare winsten en verliezen op subgroepniveau (Risk equalization 2012. An analysis of the expected profits and losses on the level of subgroups), mimeo.

  • Van Kleef, R. C., van Vliet, R. C. J. A., & van de Ven, W. P. M. M. (2011). Risicoverevening tussen zorgverzekeraars: Kwantificering modelverbeteringen 1993–2011 (risk equalization between health insurances companies: quantification of model improvements 1993–2011). Tijdschrift voor Gezondheidswetenschappen, 90, 312–326.

    Article  Google Scholar 

  • Vektis. (2006). Verzekerden mobiliteit en keuzegedrag. De feiten over de marktontwikkelingen na invoering van de basisverzekering (Mobility of insurees and choice behavior. The facts about the market developments after introduction of the basic insurance).

  • Vektis. (2007a). Zorgmonitor. Jaarboek 2007. Financiering van de zorg in 2006 (Health care monitor. Yearbook 2007. Financing health care in 2006).

  • Vektis. (2007b). Verzekerdenmobiliteit en keuzegedrag. Hoe staat het ervoor in 2007? (Mobility of insurees and choice behavior. What’s the deal in 2007?).

  • Vektis, (2008). Verzekerdenmobiliteit en keuzegedrag 2008. Begin of einde van de rust? (Mobility of insurees and choice behavior 2008. Beginning or end of the peace?).

  • Vektis. (2012). Verzekerden in beweging 2012. (Mobility among insurees 2012).

  • Vektis. (2013). Verzekerden in beweging 2013. (Mobility among insurees 2013).

  • Wolfe, J. R., & Goddeeris, J. (1991). Adverse selection, the Medigap insurance market. Journal of Health Economics, 10, 433–459.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. CPB Netherlands Bureau for Economic Policy Analysis, 80510, 2508 GM, Den Haag, The Netherlands

    Jonneke Bolhaar

  2. Department of Economics, VU University Amsterdam, De Boelelaan 1105, 1081 HV, Amsterdam, The Netherlands

    Maarten Lindeboom & Bas van der Klaauw

Authors
  1. Jonneke Bolhaar
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Maarten Lindeboom
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Bas van der Klaauw
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Bas van der Klaauw.

Additional information

We thank NIVEL and especially Judith de Jong for kindly allowing us to use data from the Health Care Consumer Panel and Peter Zuidhof from Zorgweb for supplying data on all insurance premiums for 2006. We would also like to thank participants in seminars and conferences in Aarhus, Bergen, Bristol, Mannheim, Sankt Gallen, York, EEA 2009 in Barcelona and the 18th Health Economics and Econometrics Workshop in Sardinia for useful comments.

Appendices

Appendix 1: Consumer Panel Compared to Dutch Population

In this appendix we consider how representative the Dutch Health Care Consumer Panel is. Therefore, we compare the composition to statistics on the Dutch population collected by Statistic Netherlands. Table 13 gives the results on the comparison.

Table 13 Representativeness of Dutch Health Care Consumer Panel
Full size table

Young males are under represented in the consumer panel. In the consumer panel on 21.5 % of the men are under age 45, while in the Netherlands this 46.7 %. For women the age distribution in the consumer panel closely resembles the Dutch population. Due to the difference in age composition of men in the consumer panel there is also a smaller proportion of singles without children. Furthermore, within the consumer panel a slightly higher share of men are in less than good health. The level of education is very similar. For women differences between the composition of the consumer panel and the Dutch population are less pronounced. If for men we look within age groups, the household composition and health status is very similar between the consumer panel and the Dutch population.

Appendix 2: Risk Equalization

The Dutch health insurance system includes risk equalization with both ex-ante and ex-post compensation. The goal is to make all consumers equally attractive for insurers so that risk selection is not beneficial. Ideally, ex-ante compensation would be sufficient and ex-post compensation would not be necessary. However, the calculation of the ex-ante compensation is not perfect, and therefore ex-post compensation is used to further reduce the loss insurers make on individuals that use a lot of care.

Ex-ante risk equalization (called risk equalization in the remainder) has been already introduced in 1993 in the Sickness Funds to compensate insurers for insurees with high expected health care costs. While only based on gender and age in the early 1990s, the calculation of expected health care costs has been improved over the years by adding more characteristics. In 2006, risk equalization was based on gender-age interactions (in classes of 5 years), 17 Pharmaceutical Cost Groups (PCGs),Footnote 22 14 Diagnosis Cost Groups (DCGs),Footnote 23 10 region clusters and main source of income interacted with age. The data for these calculations are the claim records supplied by insurers, supplemented with information on income source from the social insurance registers.

Van Kleef et al. (2011) calculated that ex-ante risk equalization reduced the expected loss for a broad set of subgroups where individuals with chronic conditions are over represented by 70 % in 2011. When the ex-post compensation is also taken into account, the expected loss is reduced by (in total) 78 %.

The risk equalization model has some shortcomings. For example, it ignores co-morbidity (individuals cannot be in more than one of the PCGs and DCGs), does not take into account costs for mental health care, and lacks information on characteristics like education that are known from the health economics literature to have an impact on health behaviors, health and care utilization. Indeed, Van Kleef et al. (2012) find that highly educated individuals have, on average, €400 lower health care costs per year, and after applying risk equalization still €140 remains. On highly educated women the expected profit for insurers is larger than on men.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bolhaar, J., Lindeboom, M. & van der Klaauw, B. Insurance Search, Switching Behavior and the Role of Group Contracts. De Economist 163, 25–60 (2015). https://doi.org/10.1007/s10645-014-9244-6

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10645-014-9244-6

Keywords

JEL Classification

Search

Navigation