The results show that the parameter estimate for the proportion of health care being performed
by alternative producers of health care was positive and statistically significant at the 5
percent level, indicating that a larger proportion of health care performed by alternative
producers increases economic efficiency in Swedish local county councils. The data at hand
does not, however, allow us to discriminate between the four possible motivations for this
result discussed above.
The findings in this paper coincide with the results presented by Millward and Parker (1983)
and Sandstrom and Bergstrom (2005), which have shown that competition from other
providers has increased efficiency in other sectors than healthcare in the economy. The
finding that local county councils with a larger proportion of health care being produced by
alternative producers are more efficient could, however, also be the result of production units
owned and operated by alternative producers of health care being more efficient than the
production units operated by the local county councils.
Contrary to Gerdtham et al (1999) the results also indicate that after controlling for the
proportion of health care being produced by others than the county council, the indicator
variable relating to non-socialist county councils is both negative and statistically significant
at the 5 percent level. A possible explanation for this result could be that non-socialist county
councils chose to buy a larger part of their health care from alternative health care producers
instead of investing in order to increase efficiency at their own county council facilities. This
also turns out to be the case in our data, where the non-socialist county councils have on
average 33 percent of the physician visits (divided by the number of physicians in the county
council) at production units owned and operated by alternative health care producers, while
the corresponding number for socialist county councils is 19 percent.
The results also show that there is a positive and statistically significant correlation between
having a region hospital in the county council and economic efficiency, contrary to our
expectations. Note also that there is a positive and significant correlation between our
measure of health care quality and economic efficiency. But since our measure of quality is
the 28-day mortality rate after suffering a stroke, a positive parameter estimate show that there
is a positive correlation between efficiency and mortality, indicating that there has been a
price to pay in quality for achieving high economic efficiency in the Swedish county councils.
11