Abortion as insurance

ABSTRACT This paper views abortion access as an insurance policy that protects women from unwanted pregnancies. Within this framework, we present a theoretical model where greater access provides value in the form of insurance against unwanted births and also reduces the incentive to avoid pregnancy. This model predicts that legalized abortion should lead to a reduction in the likelihood of giving birth. It also predicts that if abortion access becomes relatively inexpensive (including both monetary and psychic costs), then pregnancies would rise and births would remain unchanged or may even rise as well. We review the evidence on the impact of changes in abortion policy mainly from the United States and find support for both predictions. Then we test these hypotheses using recent changes in abortion policy in several Eastern European countries. We find that countries which changed from very restrictive to liberal abortion laws experienced a large reduction in births, highlighting the insurance value. Changes from modest restrictions to abortion available upon request, however, led to no such change in births despite large increases in abortions, indicating that pregnancies rose as well. These findings are consistent with the incentive effect implications of our model.
Phillip B. Levine and Douglas Staiger, the Department of Economics Wellesley College Dartmouth College

__________________________________________________________________________________

D. Econometric Results
Table 3 reports the results of our econometric analysis.20 In all specifications, the omitted legal status is available on request so that all coefficients for those types of abortion laws included should be interpreted relative to an on-request legal regime. For each outcome, odd-numbered columns do not include state-specific trends and even-numbered columns do include them.

1. The Abortion Rate
The first four columns of this table display our estimates for models of the abortion rate, differentiating between estimates from models using all available abortion data and from models only using those countries whose abortion data is considered to be of high quality. The results provide strong evidence that the status of abortion laws has a large impact on the abortion rate. For instance, based on all available abortion data, countries in which abortion is only legal to save the mother’s life or for specific medical reasons have abortion rates that are only about 5 percent of the level observed in countries in which abortion is legal upon request.21 Unfortunately, we cannot identify this parameter in models with the more limited, complete abortion data. Those countries/years in which abortion is available for medical and social reasons are also found to have a significantly lower abortion rate compared to those countries/years in which abortion is available upon request. In particular, when we only use those countries with complete abortion data (Columns 3 and 4), we find a robust estimate of about a 25 percent reduction in the abortion rate when these moderate restrictions are imposed.

2. Relative Female Death Rate
Columns 5 and 6 present estimates from models where the dependent variable is the log of the ratio of female-to-male death rates for those aged 15 to 44. In the model without country-specific trends, estimates indicate that the relative female death rate is 10 percent higher in countries where abortion is only available to save a mother’s life or for other specific medical reasons compared to countries in which abortion is available upon request. The introduction of country-specific trends, however, lowers this estimate and increases its standard error to the point where it is no longer statistically significant. The ambiguity in these findings makes it difficult to draw strong conclusions regarding the impact of strong abortion restrictions on maternal deaths and, hence, illegal abortion.
On the other hand, we find no evidence that countries with weaker abortion restrictions (i.e. those where abortion is available for medical or social reasons) experience any difference in relative female death rates. Point estimates are wrong-signed and statistically insignificant. This finding is important because it provides evidence that illegal abortion is probably not a tremendous problem in countries with these weaker restrictions compared to those in which abortion is available on request. It still may take place, but it seems unlikely that it is extensive or terribly risky. This conclusion will play an important role in our subsequent analysis of pregnancy behavior. (Columns 9 and 10), estimates indicate that pregnancies fall by 27 to 45 percent when abortion moves from available upon request to available only in serious medical circumstances. This estimate may be inaccurate, however, in that it fails to include counts of pregnancies that result in illegal abortions when restrictions are severe. The results from Columns 5 and 6 provide some evidence for this in that maternal mortality rises when abortion is severely restricted (in the model without country-specific trends), indicating that at least some illegal abortions are performed. This problem is identical to the one described earlier regarding the impact on pregnancies of abortion legalization in the United States. Without additional information, we cannot draw strong conclusions regarding the pregnancy impact of such a policy change. On the other hand, we are on safer ground drawing strong conclusions regarding the impact of more moderate abortion restrictions. In Columns 5 and 6 we found no evidence that maternal mortality rises when abortion is made available for medical or social reasons compared to when it is available upon request. This suggests that illegal abortions are probably not a huge problem under these circumstances. We do not conclude from this that they do not occur, only that they are not that prevalent. This is important because we also see that the pregnancy rate is estimated to fall by about 25 percent in response to these more moderate restrictions when we use all available abortion data and by roughly 10 percent when we restrict the analysis to those countries with complete abortion data. If illegal abortions are performed only infrequently in locations that have imposed these more moderate restrictions, then these estimates probably are at least close approximations of the true pregnancy effect. Based on this evidence, we conclude that pregnancies are meaningfully reduced when moderate abortion restrictions are imposed, supporting the existence of an incentive effect. ___________________________________________________

Summary of Evidence from Eastern Europe
The evidence we have presented for Eastern Europe is consistent with a growing body of recent evidence from the United States that similarly compares regions with changed abortion access to regions where it has been stable. We find that strict limits on abortion access are associated with large increases in the birth rate, on the order of 10 percent or more. Based on all available abortion numbers in these countries, we estimate that pregnancies fall by 27 to 45 percent when abortion access was very restricted (although this is most likely an over-estimate because it does not count illegal abortions). These results are somewhat larger in magnitude than estimates of the impact of abortion legalization in the United States, but consistent with earlier evidence from Romania suggesting that pregnancy rates fell about 25 percent after abortion was made illegal. In contrast, we find that modest restrictions on abortion access have no significant effects on birth rates, but do reduce abortion rates and, by implication, pregnancy rates by a substantial amount. Estimates indicate that modest restrictions on abortion access reduced abortions by about 25 percent and pregnancies by about 10 to 25 percent. Moreover, we find no evidence of a rise in maternal mortality associated with these modest restrictions, which suggests that this decline in pregnancy was not offset by any substantial rise in illegal abortions.

Our results add to the growing evidence that both pregnancy rates and the use of abortion react to changes in abortion access. If abortion access is viewed as a form of insurance, it naturally provides insurance value (in the form of avoiding unwanted births) and incentive effects (in the form of less pregnancy avoidance). Our findings suggest that both exist, and should be acknowledged in the design of abortion policy.

This article makes no attempt at evaluating the moral effects of abortion, but it proves what many of us have suspected: if abortion is viewed as legal, acceptable, and safe, the result is more activity that leads to abortion. Any thing but rare. The former Soviet States in Eastern Europe have inadvertantly proven a "laboratory" for the effects of abortion policy, with all the complications of abject poverty and wide range of religious and cultural view points. The authors are reporting on an economics study on the effects of abortion policy on the number of abortion, pregnancy and birth rates. When very strict abortion laws are first in place, the birth rate increases by 10%, but the pregnancy rate decreases by 27% to 45%. Women are able to control their sexual behavior, and that of men, in order to decrease the risk of pregnancy. Moderate restrictions cause no change in the birth rate but do decrease the abortion rates by 25% and the pregnancy rates, by 10 to 25%. When there are virtually no restrictions on abortion, or when the external and internal costs of pregnancy outweigh those of pregnancy, the birth rate may remain stable, but the pregnancy rate increases, because abortion is thought of as a fall back - insurance, if you will.

I do not think that women are victims, except in cases where they are confronted with criminals who are willing to break the law to get what they want. This report proves that women are able to control their sexual activity - and that of the men in their lives - to achieve a goal of fewer pregnancies, and that abortion on demand or with low costs to the woman result in increased "unwanted" pregnancies and abortions.

Please forgive my rudimentary formatting.

Thank you for the ping, Lady. The following from the precis clearly indicates the blindness toward considering the unborn to be individual human beings : In the model without country-specific trends, estimates indicate that the relative female death rate is 10 percent higher in countries where abortion is only available to save a mother’s life or for other specific medical reasons compared to countries in which abortion is available upon request. Obviously not a single female in the womb was counted. Reproductive rights? How about all those female little ones with developed but immature ova who were never given any choice toward 'reproductive rights' because they were slaughtered for the 'woman's choice'? It appears that the age/size of the female is all important in deciding to whom rights are extended. How very enlightened from an abortion rights feminist's standpoint, eh?

There is no attempt at any sort of judgment about the morality of abortion in this study (which could be a judgement in itself).
But,,,
the authors go on to say,
"The introduction of country-specific trends, however, lowers this estimate and increases its standard error to the point where it is no longer statistically significant. ***The ambiguity in these findings makes it difficult to draw strong conclusions regarding the impact of strong abortion restrictions on maternal deaths and, hence, illegal abortion."****

(I believe that this means that the general level of technology and health of the nation cancels any apparent changes in the (post natal) female death rate.

The truth in the article is that pregnancies, as well as abortions go down in countries that only allow abortion to save lives, and there is no *net* change in other measures including maternal death rate, infant death rate, and birth rate.

The former Soviet States in Eastern Europe have inadvertantly proven a "laboratory" for the effects of abortion policy,

Including, most importantly IMHO, the very real prospect of implosion. One thing I found interesting in Petersburg was our guide's very frank discussion of childrearing and education in Russia. The state-mandated benefits are enormous and far more substantive than the WIC and welfare carrots dangled which keep women and children bobbing somewhere just below the povery line.

While the former Soviets, like our State, has their Plans for five-year olds, they do indeed want mothers to stay home with their kids for the first two years. None of this ripping the suckling from her breast like Lady MacBeth because it's time to book it back to the office.

I think this explains the logic behind our First Family's wish to 'reduce' abortions and welcome any child "Born Alive".

They figure if the life has been Purchased, been Implanted and made it through the gauntlet that is Choice and the eugenicist March of Dimes "Health Babies Only" campaign to assure every child is vetted by amnio ... they're perfectly safe being the good cop.

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.