Reducing healthcare costs

The report is not entirely pessimistic, however, in that the efforts already made are bearing some fruit. Following the sharp rise in healthcare spending in the early 2000s, control policies were put in place to modify medical behavior and practices (e.g. the emphasis placed on risk management and medicalized control), and since 2010 the increase in healthcare spending recorded is in line with that initially voted. Nevertheless, savings (or an increase in resources) of around 2.8 billion euros a year are needed to limit growth in expenditure to 2.5% (compared with 4.4% today).

The report emphasizes the need to control spending by optimizing medical treatment paths. It is certainly possible to improve the supply of care, but it seems unlikely that this will be enough to achieve the targets set, especially as the economic crisis will lead to a reduction in health insurance resources.

Ageing, technical progress and healthcare expenditure

It is not enough to note the size and unsustainability of the health insurance deficit. It’s the result of a sharp rise in healthcare spending in the late 90s and early 2000s, whenthe amounts contributed to finance health insurance were stagnating (which shows that the problem is structural and not linked to the current crisis). But why have healthcare costs increased so much over time? And can we believe that improving the efficiency of patient care will be enough to significantly reduce these costs?

The aging of the population is one of the explanations often put forward to explain this increase in healthcare spending. The argument is based on simple arithmetic: one elderly person consumes more care than another, and the French population is aging. However, several studies deny that the  » papy boomers  » are to blame.

B. Dormont, M. Grignon and H. Huber(Health Expenditure Growth: Reassessing the Threat of Ageing) demonstrate that while healthcare expenditure rose by 54% between 1992 and 2000, only 3% of this increase is attributable to the ageing population. A priori counter-intuitively, it is technical progress that accounts for the bulk of this increase in healthcare spending: changes in medical practices have led to a 58% rise in healthcare expenditure (1).

Technical progress in the medical field not only reduces the cost of various interventions, but also, and above all, facilitates access to care for people who were previously deprived of it. A paradox then arises: it is because patients are better cared for, with more efficient medical techniques, that the deficit increases.

The ONDAM is right to aim for productive efficiency in our healthcare systems: waste is socially harmful, but a real quest for efficiency in our system (if it is not a pretext for rationing access to care) could ultimately have the opposite effect on deficits: more people would be better treated at a lower unit cost of treatment. There is no guarantee that the total cost of our healthcare system (number of patients x unit cost) will also fall. The volume effect, the increase in the number of patients induced by better practices, could outweigh the reduction in costs.

If the problem is not demographic, a mechanical solution is unthinkable. In this case, we need to reflect on the issue and work together to strike the right balance between justice and economic sustainability.

More specifically, we need to ask whether society values its health sufficiently to agree to fund its healthcare system. We also need to ask whether companies can make a sufficient contribution to the healthcare system. What would be the consequences for our societies of an increasing de-reimbursement of healthcare expenditure, for example through an increase in co-payments or a limitation of the sphere of reimbursed care?

Are individuals prepared to pay for health insurance?

As B.Dormont explains, when asked, people say they are willing to pay a significant proportion of their income to live longer and in good health. Once this first milestone has been reached, methodological difficulties make it almost impossible to assess individuals’ willingness to pay for the healthcare system, i.e. the total amount of their budget they are prepared to allocate to financing health insurance.

Citing Cutler’s work, B. Dormont explains that a (conservative) value for an additional year of life would be around $100,000 in the United States. Taking this value into account, the overwhelming majority of new care techniques would generate substantial profits.

The example of heart attacks is instructive. New treatment techniques can increase life expectancy by around a year for people who have suffered a heart attack. Subtracting a cost of life of $30,000 for that same year, we obtain a residual value seven times greater than the cost of treatment ($70,000 for a treatment cost of $10,000).

Using once again the « statistical value of life » approach (this value is obtained by comparing the amount an individual is prepared to spend for a marginal reduction in risk), it is also possible to estimate what the increase in GDP would be if gains in life expectancy were integrated with the value attributed to them by individuals (2). Murphy and Topel (2006) have thus estimated that the annual value of healthcare expenditure in the USA would be 32% of GDP, a value far higher than the 15% of GDP devoted by the USA to healthcare expenditure.

Society as a whole places a high value on healthcare spending. But what about companies? The French health insurance system is unique in that it is largely financed by employer contributions. The proportion of employer contributions dedicated to « Sickness, maternity, disability, death, solidarity » amounts to 13.10% of the total salary paid. This raises the question of whether companies receive compensation in the form of productivity gains.

Productivity and health insurance

The question of productivity gains, or losses, caused by health insurance is essential for two reasons. Firstly, it raises the question of corporate participation in financing the healthcare system. Secondly, it allows us to measure the social gain of health insurance from a different angle than that of the « statistical value of life », a concept that is obscure to non-specialists and may seem cynical to the general public. It is no longer a question of measuring the value individuals place on their health, but of direct productive gains.

However, studies carried out on this subject in France are either rare or non-existent, and we have to turn to American and Canadian work to study the effects of health insurance on worker productivity.

A first fact observed by Guy Lacroix seems to suggest a simple answer to this question: the more generous a healthcare system is (in terms of days of absenteeism fully compensated by health insurance), the more likely workers are to be absent. On the contrary, when policies were tightened in Germany or Scandinavia, a decrease in absenteeism was observed. This suggests « abuse » on the part of certain workers who take advantage of a generous system, resulting in a loss of productivity induced by health insurance. Productivity gains would therefore require a less generous system.

However, American studies (e.g.: Health and Productivity Among U.S. Workers by Karen Davis et al. 2005) show that the main problem facing companies is not the absenteeism of sick workers, but rather their « presenteeism ». Workers without health insurance and without the possibility of resting for a few days at home see their productivity drop drastically for longer periods than others who are able to rest, without seeing their income drop appreciably. In addition, they are much more prone to « long-term » illnesses, which prevent them from returning to work in the long term. Finally, they exert a negative externality on their colleagues, who are more likely to fall ill in their turn.

To illustrate the problem of presenteeism, Ron Goetzel et al estimate that of the three to four hundred dollars an employer spends each year per employee on hypertension, heart problems, depression and arthritis, only 10 to 20% is explained by worker absenteeism, while 18 to 60% of this expenditure is explained by employee presenteeism.

Another article,  » Health insurance as a productive factor », by Allan Dizioli and Roberto Pinheiro, stands in more direct opposition to Guy Lacroix’s, showing that in the USA, employees with health insurance are 52% less absent than their non-covered counterparts (representing 2-3 working days per year). Furthermore, a major advantage of this article is that it proposes an analytical framework for studying changes in regulations, and in particular the case where providing workers with health insurance becomes compulsory for companies in the USA. Under this hypothesis, we would see an increase in labor productivity (and a decrease in the number of sick workers), a decrease in unemployment, an increase in total output, a decrease in wage differentials between large and small companies, but also a decrease in aggregate corporate profit.

Of course, the model presented here is not completely adapted to the French case, where health insurance is compulsory and largely centralized, with contributions linked to the employee’s salary. It does, however, offer some food for thought. Indeed, given the structural deficit of social security and health insurance, it seems tempting to reduce the publicshare of health insurance, leaving greater latitude to private insurance and supplementary schemes contracted directly by individuals or negotiated by companies.

In this case, a re-use of the Dizioli and Pinheiro model predicts a decrease in gross domestic product, an increase in unemployment and in the number of sick individuals in society, and finally, an even greater polarization between large companies, able to offer high salaries and health insurance, and small ones, unable to provide the same benefits to workers.

So, while it would benefit companies, a reduction in social security contributions, combined with a reduction in health insurance benefits, could be detrimental to French society (higher unemployment and lower GDP).

Conclusion

It appears that our societies place a high value on health, and therefore on health insurance, through two mechanisms. On the one hand, people like to be healthy. On the other hand, healthy people work better.

This is why the debate on health insurance financing should not focus exclusively on issues of cost reduction and optimizing healthcare pathways: these policies will have only a limited effect on deficits, and may even, paradoxically, contribute to increasing them. Moreover, they are often justified in order to reduce the scope of health insurance, which is socially harmful.

The real problem is to define who should finance this system, and in what proportion. Thanks to health insurance, companies benefit from substantial productivity gains. However, compulsory health insurance reduces profits, especially for small businesses. This problem of incentives is probably exacerbated by the specific nature of the French system, which greatly increases payroll costs.

It is probably necessary to transfer part of this cost to other players, but it seems unfair to pass it on to the user alone, which could lead to a drop in the number of people benefiting from our healthcare system, and thus to a contraction in labor productivity and economic activity.

References

Inspection générale des finances (2012): « Rapport pour la maîtrise de l’ONDAM 2013-
2017 « .
B.Dormont (2009), « Les dépenses de Santé, une augmentation salutaire? », Collection du
CEPREMAP
B.Dormont (2010), « Le vieillissement ne fera pas exploser les dépenses de santé », Revue Esprit, July 2010
Ellen O’Brien (2003), « Employer Benefits from Workers’ health insurance », Milbank Quarterly
Volume 81, Issue 1, pages 5-43, March 2003
K. Murphy, R.H. Topel (2006): « The value of Life and Longevity », Journal of Political Economy, 2006, vol114,n°5, p871-904.
Sang V. Nguyen, Alice M. Zawacki (2009): « Health Insurance and productivity: Evidence from the manufacturing sector », US Census Bureau Center for Economic Studies Paper No. CES-WP- 09-27, September 1, 2009