Ponnuru on voting and abortion
I hope that readers (and others) who have read Matthew Boudway’s open letter to Deal Hudson (here) will also read Ramesh Ponnuru’s recent piece, “Conscientious Voting,” here. A taste:
Feuerherd asks, “[I]s it fair for a Catholic like me to suspect that the liberal economic policies of the Democratic candidate, whether Obama or Sen. Hillary Rodham Clinton, will result in less dire poverty and thus perhaps fewer abortions? And isn’t that supposed to be the goal?” Anyone who wants to cast a ballot on this assumption has a moral obligation to investigate whether it is, in fact, true that 1) Democratic policies would reduce poverty much more than Republican ones would and 2) that abortion and poverty rates correlate in as straightforward a manner as Feuerherd idly (and conveniently) supposes. I am not aware of research that corroborates point two, let alone both of them taken together.
And there is another problem with this argument, which is that a reduction in the number of abortions is not the only goal that pro-lifers should have. Also important is that the law stop treating unborn children as subhuman creatures who may legitimately be denied the protections of the law against unjust killing. Obama himself may be perfectly sincere in willing that fewer women exercise the (supposed) right to abortion even while he supports keeping that option legal and making it subsidized. I have no reason to doubt that he is. But he also wills that unborn children be denied the basic legal protection from homicide that you and I enjoy. The Catholic Church wants voters to take that injustice seriously; more seriously than Feuerherd seems inclined to take it. But of course it cannot (and has no ambition to) force any voter to do anything.
Comments are closed.
President Bush has, for good or ill, spent much effort and energy making a case for the “war on terror” and the war in Iraq. He and his administration have devoted countless speeches to the task of persuading the American public of the necessity and justice of past, present, and intended future actions of war. No matter what his approval ratings, he has stuck to his message with passion and persistency. I haven’t done the math, but I suspect he’s spent most of his time speaking to the public in the the act of selling his war.
Here’s my question: Why have we not seen this same passion for persuading the American public about the evils of abortion from our president or other pro-life politicians? I’ve heard it said that the president affects the abortion issue most by the judges he appoints, but that may not be the case. The president has a podium from which to speak daily to the American people. He should use it.
Could it be that reducing abortion and establishing legal protections for the unborn are not priorities for many of our prominent pro-life politicians?
As for the link between abortion and poverty, I did a basic investigation last year” http://vox-nova.com/2007/05/22/abortion-and-economics/. There is a very clear link.
Dr. Michael New has researched the affect of state legislation on abortion, showing a very clear link.
http://www.heritage.org/Research/Family/cda07-01.cfm
MM,
Using your approach to data analysis in that post, the greatest directional improvement was under Reagan, since up until he took office the abortion rate was sky-rocketing, and under Reagan it came to an immediate standstill and then began to slowly decline.
Was there an incredible upsurge in the number of individuals and families living under the poverty line in the Ford and Carter years which Reagan stopped? Also, do your correlations hold up at all if you extend the economic and abortion data into Bush’s second term?
The higher rates of abortion among the economically disadvantaged probably have a lot to do with much higher un-planned pregnancy rates among the economically disadvantaged. (Statistically, birth control is much, much less effective on the poor.) What would be very interesting would be to know the percentage of unplanned pregnancies that end in abortion across income bands. It wouldn’t surprise me if the poor end unplanned pregnancies in abortion at the same or a lower rate than the rich.
“Was there an incredible upsurge in the number of individuals and families living under the poverty line in the Ford and Carter years which Reagan stopped?”
It could be perhaps that certain changes in the legal climate impacted the rise in abortion rates during those years. I think I remember learning in Con Law about a decision that overturned the abortion restrictions of 46 states…but then again, it’s probably just poverty…hey don’t the Democrats want to help the poor, whereas Republicans want to keep them that way…and didn’t Democrats cause the dotcom boom thereby reducing poverty….I should probably vote for Democrats…’yes we can’ ensure abortion stays ‘safe, legal, and rare’.
MM — That post strikes me as extremely irresponsible. Given that you obviously know what an OLS regression is, you also must know that what you did in that post was hogwash. I’ve never seen a legitimate social scientist who purported to identify a causal effect just by running a simple OLS regression with so few data points and with absolutely no independent variables other than the supposed causal factor. That post might serve to bamboozle people who don’t know anything about econometrics, and who think that if it’s in an equation it must be true, but it’s not much good for anything else.
MM — That post strikes me as extremely irresponsible. Given that you obviously know what an OLS regression is, you also must know that what you did in that post was hogwash. I’ve never seen a legitimate social scientist who purported to identify a causal effect just by running a simple OLS regression with so few data points and with absolutely no independent variables other than the supposed causal factor. That post might serve to bamboozle people who don’t know anything about econometrics, and who think that any equation must be true, but it’s not much good for anything else.
Stated more directly, my objection to MM’s study are the following:
1) The post all but ignores the largest factor impacting the abortion rate in the study; namely, the removal of any and all abortion restrictions.
2) I believe poverty and abortion are linked, however I am far from convinced that Clinton was responsible for the economic prosperity and ‘irrational exuberance’ that led to the healthy economy and corresponding reduction in poverty in the 1990′s.
3) I am curious how the rates of the past four years would impact the study.
4) The study and the ensuing commentary seem calculated to suggest a certain approach to the evaluation of the political parties that omits what, for me at least, is one of the most important considerations. Namely, the fact that one of the parties will not tolerate deviation from a pro-abortion platform in any of its candidates.
Just as a note for the moderators, the comments do not seem to be posting sequentially…
Stuart and others: I not not claim to purport a causal relationship. All I am saying is that the data suggest a strong correlation. Is there an endogeneity issue? Possibly. But what omitted variables are biasing the estimate? The intent here is not to get published in the AER. It is to show that there is a clear relationship between economic conditions and abortion (and by the way, this evidence says nothing about what would happen if greater restrictions came into place– I think that would be obvious). It is to attack the myth that somehow the nominally pro-life Republicans are better for abortion.
And Darwin: the post-Roe decline started under Reagan, sure. But you need to exaplain why the abortion rate declined by an average 0.3 percentage points a year under Reagan, Bush I and Bush II, and 0.5 percentage points a year under Clinton. It seems quite obvious that economics matters, especially since poverty rates declined under Clinton bur rose again under Bush.
I don’t think the assumption that the economy improved because Clinton was president is necessarily valid. I do know that voting for someone who will continue to entrench legal abortion in American politics is a bad idea if you want to end abortion. I also know that, as fetuses (feti?) are human beings, any law to keep abortion legal is a blatant disregard of their lives as well as a form of unjust discrimination. In my mind, one must ask the question, “If I were alive in the 1960′s, would I be willing to vote for a (hypothetical) candidate who would strive to legally uphold segregation since his economic policies might improve the economy and lessen racial tension and personal prejudice?” I don’t imagine anyone answering “yes” to that question, as we all agree racism is wrong.
The economics here are a red herring. A tool of the pro-abortionists and those who wish to be blind to abortion in their voting preference to justify their choices. The critical point espoused in every single Church document is that the goal is to eliminate the injustice that denies basic human rights to a class based on their developmental level.
One could make an argument that African slaves fared much better than those left behind in Africa which was by no means some sort of Utopia. Those arguments would be of no value either.
As Kevin suggests, it there may be a strong correlation between the decline of the black family and the decline of segregation. Does that mean we could ignore the fundamental injustice of segregation? Of course not.
God Bless,
Matt
But you need to exaplain why the abortion rate declined by an average 0.3 percentage points a year under Reagan, Bush I and Bush II, and 0.5 percentage points a year under Clinton. It seems quite obvious that economics matters, especially since poverty rates declined under Clinton bur rose again under Bush.
I’ll have to go dig up the data on this again, but I recall reading a few months ago (for some reason I was thinking it was from one of the conservative authors here, but it must have been elsewhere) that when you re-run the abortion rate numbers for the first six years of Bush’s presidency (for which we now have data, but didn’t when this abortion-economics trope first showed up) rather than the first three, it turns out that abortion decreased at the same rate during his presidency as a whole as under Clinton’s as a whole.
Darwin– I ran the latest numbers. Bush II certainly looks better when you take it through 2005, but he merely goes from standstill to the same average annual decline as his father. I concede that when we get data through 2008, it might look better, but not so far.
Stuart and others: I not not claim to purport a causal relationship. All I am saying is that the data suggest a strong correlation.
Your original post said, “But it seems clear that economic factors, insofar as they affect poverty, affect abortion patterns,” and even now you’re trying to “show that there is a clear relationship between economic conditions and abortion.” To the average reader, you’re claiming causality.
But you haven’t even shown correlation yet. For one thing, you don’t seem to take into account the drastic rise in abortion during the early to mid 1970s. The abortion rate nearly doubled in just a handful of years, which is a rate of change that far exceeds any changes that have happened since. But did poverty rise at the same time? No. That suggests that the legal change wrought in Roe v. Wade could have had vastly more impact on the abortion rate than poverty did. I wonder why you exclude those years; the term cherry-picking comes to mind.
But what omitted variables are biasing the estimate?
You’ve got to be kidding. You know what OLS is and you seem to know how to use Stata or something, but you aren’t aware that there might be an omitted variable problem when you didn’t control for anything whatsoever? Any real effort at studying this issue would control for changes in religious attitudes, demographics, the average price of abortion, various legal restrictions passed in the 1990s (which do have an effect: see, e.g., http://findarticles.com/p/articles/mi_qa3634/is_199803/ai_n8785342/print ), availability of state subsidies, levels of marriage and labor force participation (which affect women’s willingness to get abortion), and the like.
Other things to look at: the availability and price of various forms of contraception, and the proportion of childbearing women who were in their teens or 20s (if younger people are more likely to get abortions, then the rate will change if the population ages).
So there are at least 9 omitted variables that I can think of within a few minutes.
Stuart: I was asking you what variables you thought might be omitted, not questioning the nature of the problem. But since you insist on being an arrogant jackass (lawyer, right?), I’ll give you my qualifications: Ph.D. in economics, with plenty of empirical experience.
Of the variables you list, I will concede that legal restrictions could have an effect, but since my simple analysis was based on the US aggregate, it couldn’t be included. Try a state-level panel if you feel up to it. Religious attitudes? Interesting to see if you could get an empirical proxy for evangelicals changing their minds (ha!). As for other economic variables, I would put forward median per capita income and income distribution. Also educational attainment (which could work against the partricipation channel you propose). Availability and cost of health care. And yes, demographics. So, yes, it’s complicated.
But, as I keep saying, the largest decline came under Clinton, and whether you credit him or not, that correlates with the increase in broad-based growth at the time– the only time since the 1970s when productivity not only grew robustly, but was broadly shared. I think the poverty rate stands out as a good proxy for what happens during that period which is why I honed in on this variable.
Make that 10: Availability of abortion clinics, given that higher travel costs seem to affect the abortion rate.
At some point you might worry about multicollinearity, of course, but you aren’t in any danger of that!
Another interesting way to do this would be to compare abortion rates across different states. See, e.g., http://www.statemaster.com/graph/hea_abo_rat-health-abortion-rate
But the problem (at least the problem for you) is this: First, if you look at a nationwide chart of abortion rates, some of the richest states have high abortion rates (i.e., in the Northeast) while some of the poorest states (like Mississippi, West Virginia, and Kentucky) have the lowest rates. http://www.statemaster.com/graph/hea_abo_rat-health-abortion-rate Second, one study a while back found that abortion was less common in poorer states (this was evidently a real regression that controlled for other factors). See http://www.popline.org/docs/1076/095120.html
So if you wanted to manipulate the abortion rate via public policy, you should probably get to work trying to transform everyone into impoverished subsistence farmers in the Bible Belt.
Morning’s Minion,
Doesn’t having a Ph.D in economics make these omissions worse, not better? If someone made a basic error about the law, the fact that he had a law degree would make the error more glaring, not less so.
By the way, could you dig up a real study that examines changes in poverty and the abortion rate? I’d be interested to see it.
So if you wanted to manipulate the abortion rate via public policy, you should probably get to work trying to transform everyone into impoverished subsistence farmers in the Bible Belt.
Which might not be a bad idea for other reasons as well; carbon emissions would probably go way down.
But, as I keep saying, the largest decline came under Clinton, and whether you credit him or not, that correlates with the increase in broad-based growth at the time
And as I keep saying, a far more drastic increase came after 1973, and does NOT correspond to an increase in poverty. Yet you omitted those years even from the barebones analysis. Why?
In my “monkey regression” (to use the phrase of a colleague), the starting point is 1973, as that’s when the Guttmacher data start. And if I had found a “real” study, I would have linked to it, rather than having to do one myself.
Your post said, “In particular, let’s do a simple ordinary least squares regression of the abortion rate and the abortion ratio on a constant plus the poverty rate, 1980-2003.” That indicated to me that you were selecting 1980 as the start date, not 1973. Am I misunderstanding something there, or was that a typo, or what?
I guess I was doing 1980 back then, for reasons I can’t recall. But I just re-ran the numbers from 1973– no change. You need to think of 1973 as a major regime change. Once in the new regime, economics definitely affects abortion rates.
No change, even with several years of drastic increases in abortion but no change in poverty rate? I can’t figure out how that would be true.
In any event, you seem to admit that a valid analysis would control for a dozen other factors, so I’m not sure why it would be useful to find a raw correlation.
MM,
You need to think of 1973 as a major regime change. Once in the new regime, economics definitely affects abortion rates.
Precisely the whole point. What was the major regime change? The legalized murder of the unborn. Make it illegal again, and poverty notwithstanding it will decline back to obscurity again.
Why don’t you re-look at the Church’s teachings on abortion, it is clear that the very legality of abortion is the gravest of evils.
God Bless,
Matt
Doesn’t having a Ph.D in economics make these omissions worse, not better?
Exactly — having a PhD means that MM should have known better than to use technical-sounding lingo that could easily fool the average reader into thinking that the analysis was at all meaningful. What MM did was precisely on the level of a Republican “economist” who produces an equation purporting to prove that all tax cuts cause revenue growth, by regressing federal income tax revenue from 2000-2008 (unadjusted for inflation) on the top marginal tax rate, without taking any other factors into account.
That’s ridiculous, and you know it. The regression you describe is ludicrious. My analysis is simple, univariate, but sensible. Even without resorting to time series analysis, it is hard to deny that economics matters for abortion, and that poverty is the most pertinent factor. The underlying Guttmacher data show this clearly. Had I made it more complicated, you would have come up with other objections– because this kind of reasoning touches a sacred cow of so many on the right. Well, I’m going to slay those sacred cows. Get over it!
No, my example is not ridiculous. Like your analysis of the abortion/poverty issue, it would pretend that something meaningful could be gleaned from a regression that had only a few data points and that controlled for nothing whatsoever. Such an analysis is either incompetent or dishonest. As was your post.
In addition, you say that “poverty is the most pertinent factor.” Really? Where do you get this?
Had I made it more complicated, you would have come up with other objections
Possibly, depending on how you made it more complicated (i.e., if you cherry-picked variables that increased the effect size while ignoring any other variables). Nonetheless, that’s a pathetic excuse for failing to control for anything.
My guess is that the relationship between the poverty rate and the abortion rate is more complicated than Morning’s Minion suggests. Children are a financial burden, so we might conclude that the poorer a woman is, the less likely she will feel able to afford this burden, and the more likely she will seek an abortion. But this ignores opportunity costs. A 20 year old college student or a 30 year old young professional has more to lose from having a child than would a 20 or 30 year old high school drop out. So I don’t think we can say, in the absence of any data, what the relationship there is between economic condition and the abortion rate.