This post is about single-parent families and the role of welfare. Like the last post, it was inspired by a comment on an earlier entry, and aims to debunk a few common myths.

A regular (indeed, prolific) commenter here made the following observation this morning:

I think the welfare programs of the 60’s encouraged fathers to leave/not join their families==women got more money without men around to spend the welfare check on beer which led to being able to live on welfare by having kids.

I’m not unsympathetic to the idea that government assistance should not come with strings which interfere with healthy family relationships. However, it’s apparent that there are several myths about black families, welfare, and single-parent households which are entrenched in this country.

I have previously posted about the persistent myth that most welfare recipients in this country are black. It is true, of course, that black families are disproportionately represented on the welfare rolls, if only because blacks are disproportionately represented among poorer Americans. However, depending on how welfare is defined, either whites are the majority of welfare recipients or whites and blacks are represented in roughly equal numbers.

Welfare incentives for having children

Does welfare encourage parents to have children in order to receive welfare checks?

The answer is no, and this is not merely a result of welfare reform in the mid-1990s, which aimed to eliminate such incentives. Between 1969 and 1994, for instance, the average size of families on welfare decreased significantly, from four people to less than three. As of 1994, in fact, the average family on welfare was no larger than the average family not on welfare; 43% of welfare families included just one child, while 73% consisted of one or two children. Research even before welfare reform showed that welfare benefits were not a significant incentive to have children.

Welfare incentives for how to raise children

Does welfare encourage women to have children without a father in the picture?

The issue of single parents is often described as a problem of unwed mothers and absent fathers, particularly when describing black parents or families on welfare. However, even in these cases, this portrait is strikingly misleading.

Single-parenting is a phenomenon which disproportionately affects black families, but which has significantly impacted families of all races since the early 1960s. The trend towards unwed parents, in fact, arose simultaneously in families both on and off welfare, as a result of the social revolution of the 1960s and 1970s, not because of the Great Society programs of the 1960s.

According to a 2001 Brookings Institution paper on the Fragile Families Study, unmarried parents are often parents who simply aren’t married. Half of unmarried parents are living together when their child is born, and another third of such couples are in a romantic relationship at the time. In only 9% of cases, in fact, do the parents have little or no contact.

Even more strikingly, four-fifths of the unwed fathers in this study provided financial support during the pregnancy, and when born, four-fifths of the children were given their father’s last name. Most of the parents, both mothers and fathers, say that they want both of them to be involved in raising the child. Almost three-quarters of the mothers, in fact, believed that their chances of marrying the father were even or better. Almost two-thirds said that they believe that being married is better for the children.

According to the same Brookings Institution paper, the problem with unwed parents isn’t welfare at all. On the contrary, it’s poverty, education, and lack of opportunities. While almost all fathers in the study, for instance, had worked in the year prior to the birth of their children, almost 30% were out of work in the week before the birth. Similarly, unwed parents were much more likely not to have graduated high school. On the other hand, unwed parents were highly committed to their children and to each other.

The means-testing of welfare does mean that parents have an incentive to live apart, or to convince the government that they do.

The effects of single-parent families

What are the effects of having only one parent raising a child?

A 2003 study in the journal Child Development found that in black families, the presence or absence of a father had no effect on the child’s cognitive abilities, language competence, or behavior. However, the quality of the role played by fathers in the family played a significant role in all of these measures of the well-being of their child. This suggests that the role of a father can be important in child rearing, but that merely encouraging marriage, joint living arrangements, or paternal involvement is not the answer.

More extensive research on single-parenting has been conducted by Sara S. McLanahan, a scholar at Princeton, some of which is summarized in her article, “Father Absence and the Welfare of Children,” in Eileen Mavis Hetherington, ed., Coping with Divorce, Single Parenting, and Remarriage: A Risk and Resiliency Perspective (1999).

McLanahan finds that “most children who grow up with single parents do quite well.” She does find certain modest, but not large, differences between children raised by one or two parents, in terms of completing high school, going to college, and being steadily employed. But she concludes that there is not enough difference to suggest that absent fathers are the major cause of significant social ills.

It’s also important, I think, to pay attention to why McLanahan detects some differences in children raised with one parent, usually the mother, rather than two. She finds that roughly half of the difference is attributable to the lack of the second parent’s income. The other half is split roughly equally between diminished parental attention and disruptions caused by residential mobility and, thus, disrupted neighborhood and community ties. Notable is what is absent from these causal factors, including any distinctive characteristic of fathers as parents, moral character to two-parent families, or other factors which might not be compensated for in other ways.

35 Responses to “Welfare, unwed mothers, and absent fathers”


  1. bobbo says:

    Prolific huh? Well, a quick review of the site reveals I am exceeded only by YOU. I wonder if we are driven by the very same thing? Probably not. I accept you are driven by a desire to educate and inform. For me, that would be a secondary outcome.

    Well, before I revert to the statistical norm of zero, let me continue my profligate ways:

    You say: "Does welfare encourage parents to have children in order to receive welfare checks?

    The answer is no," /// So, providing government funding of an activity does not encourage it. Good to know. This bit of economic theory would be a divining rod for identifying bias.

    but what is funny is the contradictory conclusions/statements made right after this statement:

    1. and this is not merely a result of welfare reform in the mid-1990s, which aimed to eliminate such incentives //// This is an express admission that welfare did incentivize-at least prior to the mid 90's, and indeed my time reference was to the 60-70's. While each state can have its own rules, in the 60's in California if there was a "man in the house" you didn't get welfare (or more specifically AFDC-Aid to Families (sic) with Dependent Children). I think the "rule" was that such status was not disqualifying, but in practice the welfare department did not approve the funds while it was "researching" the male in the houses status re fatherhood and income support so as to prevent "fraud." More than an incentive–a requirement to get welfare.

    Well, I would go on, but my router is acting up and I can't link to supportive websites. The study continues its doublespeak though in quite a dramatic fashion. Don't often see that "in print." Good example to save for cases of bias on display.


  2. James says:

    So, providing government funding of an activity does not encourage it. Good to know.

    The issue isn’t whether state-based financial incentives can encourage an activity.

    The issue is whether people will have children merely because of financial incentives to do so, and whether welfare provides such incentives, once the costs of having children are considered.

    I’m not surprised, frankly, that research consistently shows that Americans do not bear children in order to receive welfare benefits.

    This is an express admission that welfare did incentivize-at least prior to the mid 90’s

    No, bobbo, this is an admission that prior to 1996, politicians frequently claimed that welfare provided such incentives. That’s why I wrote that welfare reform “aimed” to eliminate such incentives.

    In fact, of course, welfare benefits are generally going to be larger with each child. It would be perverse if that weren’t the case. The issue, again, is whether those benefits are so large that parents will actually have more money, despite the costs of having a child, and whether any net benefits, which would have to be quite small, would cause people to have children for those benefits.

    indeed my time reference was to the 60-70’s.

    You did refer to “the welfare programs of the 60’s.” But as you can see, the research indicates that welfare did not cause recipients to have children, even between the 1960s and 1990s.

    While each state can have its own rules, in the 60’s in California if there was a “man in the house” you didn’t get welfare

    Which naturally created incentives, I agree, in order to receive welfare. For instance, as I noted in the post, such rules encouraged parents to live apart (or to say that they do) even if they’re both fully engaged parents, and even if they pool income for the support of their children.

    The study continues its doublespeak though in quite a dramatic fashion.

    I look forward to hearing more from you about what you think is “doublespeak” in the study in question, once your router troubles are cleared up.


  3. bobbo says:

    James. I laugh at myself. I often criticize people who use their "personal experience" as the basis to review/opine on societal issues. I have avoided stating the main basis for my opinion because it is only that, my personal exposure:

    1. My Dad was a Senior Welfare Case Worker "3" in California in the 60's. He worked exclusively with 2 and 3rd generation welfare recipients and their children who were going to be the 4th generation. He would occasionally talk about welfare in Calif for his group: mainly they are trapped. Bad Health. Bad Attitudes. No education. No books. Having kids to qualify for welfare being the "goal."

    Later I got involved in some low rent housing. Single mothers raising daughters to go on welfare by having kiddies.

    Did these trapped folks have kids "merely to recieve welfare?" Yes they did. They had no other perceived options, often they had no other better options in fact once you add in healthcare coverage that goes with the package. Had to make well above minimum wage to match the income available.

    So, I won't use my personal experience to say what most Americans are incentivized to do, but I will say that "some" Americans were expressly motivated by that interest in the 60-70's.

    Just read a story about some guy who had an industrial accident and lost a finger. He got workers comp recovery for it. Shortly thereafter he lost another finger. Then a third. Not everyone is motivated like everyone else. Its how we are all the same yet different at the same time.

    Since you asked, here is what I read that made me think "doublespeak":

    From the pdf (above Table 2): "(The findings) support the argument that a substantial proportion of unwed parents are not ready for marriage because of low employment skills….."

    The Paragraph starting with "TANF Reforms" restates the still present disincentives for family formation saying "Although many state TANF programs appear to have reduced or eliminated restrictions for two-parent families, others still retain such restrictions." Thats about as wiffle waffle as you can get.

    Likewise later with their main recommendation: "Although a public child support benefit would also increase government expenditures and parents' incentives for living apart, it would nonetheless reduce the poverty, insecurity, and welfare dependence of single mothers and their children."

    Book tv about a year ago had a social policy guy recommend a welfare program where "everyone" got around $20K per year without further restrictions. Immediately, he said, black families would form because "as a unit" a couple would have money–the guys could not steal the womans money claiming they had none. Men would be seen as valuable assets instead of liabilities. Incentives in place not to have kiddies. He had cost figures what with trades offs in welfare, crime etc made the program seem like a real plus.

    Creative thinking is always a bit odd at first blush.


  4. James says:

    Bobbo, I appreciate that both you and your father believe you were interacting with welfare recipients who were having children "merely to receive welfare," and I certainly can't deny that it might be true.

    Personal experience can be tricky, however. In this case, how did your father know that his welfare cases were having children to qualify for welfare? How do you know that the single mothers you knew through low-incoming housing were raising their daughters to have children for the welfare checks?

    The most obvious possibility is that some of these people were saying exactly that. In fact, I'm not sure how else you would think you knew. How, though, do we know what statements we can trust? People sometimes make statements that aren't true, for a whole host of reasons. Other times, they may believe what they're saying, but not understand their own motivations sufficiently.

    I think it's entirely possible that a young, single woman could have children for a host of reasons, including several of the bad reasons that your father identified, while saying or believing that she was acting for another reason.

    This is why I believe it's important to rely, as much as possible, on objective research which is conducted with controls to identify what behavior is actually occurring, and why. If research studies with large numbers of subjects show that welfare incentives do not, in fact, result in recipients having more children, then these studies are worth much more than a handful of anecdotal conversations with welfare recipients.

    You are, of course, quite right that individuals are motivated differently, and that there may be some welfare recipients who respond to incentives differently than most.

    I'm afraid you're still losing me on the "doublespeak" issue. For instance, I'm not sure what you see as "wiffle waffle" about reporting, accurately, that some state have reduced or eliminated disincentives for two-parent families, while others have not. That's a description of state welfare requirements, and doesn't affect the paper's conclusions at all.

    Likewise, you quote their conclusion, but you don't say what you think amounts to "doublespeak." Haven't they carefully identified the trade offs involved (more spending and perhaps parents living apart; less poverty, insecurity, and long-term welfare dependence)? "Doublespeak" is usually taken to mean language which obscures its meaning or distorts the truth.


  5. bobbo says:

    I admit my own interaction was limited and biased in a number of ways. But dear old Dad was "an experienced specialist" in the field. Who you gonna believe, an academic with a clipboard or a specialist who has maintained monthly contact with a recipient family over the generations? ((Plus self reporting?)) Ok–he was biased and not objective by definition. ((I indicated his biased exposure by virture of his Level 3 caseload==the multi-generational recipients. Just the type that would have such twisted values? Not the general recipient at all.))

    I'm embarrassed to say I can't support my claim of wiffle-waffle. I know what I thought I read, but going back to pull the quotes from the studies, I can't find the language even after I downloaded the pdf and converted it to a word document and used word search functions. Perhaps it was from one of the supporting or later searched links?

    I agree in general that "some" people will do just about anything and that societal programs should be based on something with more general applicability. To that issue, it would be the rare situation where people aren't motivated by money==regardless of whatever other issues may be attached or consequential. A few studies wouldn't budge me from that conviction without some sort of paradigm shifting explanation. Being unfair to their own kiddies doesn't create enough of an inhibition in my view. Look at OctoMom?


  6. James says:

    I think you're right to emphasize that your father worked with a special subset of the general welfare population. It's possible that there was such an effect among that group of recipients, which is undetected in larger studies. However, we would certainly expect to see some effect, if that group was as affected by welfare incentives as your father suspected they were.

    I'm not sure, either, where you found that language. It must have been somewhere in those links, though, since I know you found it somewhere. :-)

    You seem to assume that welfare recipients will generally be motivated by money to have children for the extra benefits. I think there are two problems with this assumption: First, it simply doesn't square with the known facts. You say that "a few studies" won't sway you from your belief about human nature, but I'm inclined to believe what people actually do, before your musings or mine about how they might make decisions. Second, your belief assumes that the benefits for having a child are large enough that, even subtracting the considerable expense of raising a child, there will be money left over–and not pocket change, but enough to motivate someone to go through the process having pregnancy and childbirth, as well as raising a child for decades.

    To me, it's clear why the evidence indicates that having a child for the welfare benefits is the rare occurrence.


  7. bobbo says:

    James, I've been thinking about this as well. I think you have me. (Rats!!) Actually, we BOTH have me for as I said, I hesitate to state the basis for my opinion==personal and anecdotal, not scientific and subjected to rigorous protocols. I also think "the law" has changed since Dad was assigned these problem cases. But before I post this to you, I'll spend just a little time on the google.

    I googled (welfare incentive to have children) and the 2-3rd post was the following claiming that women on welfare have 3 times the birthrate of women not on welfare.

    Is there anything "tricky" in the wording of these two different studies that seem to be contradicting each other on what should be objective facts?

    http://mjperry.blogspot.com/2008/08/if-you-subsid


  8. James says:

    Thanks once again, bobbo, for having such a positive attitude about these discussions.

    It's true that the Census paper in question reached that conclusion. However, I don't see any contradiction here.

    First of all, the paper doesn't suggest that women on welfare have three times the birthrate. Just that they were three times as likely to have had a child in the last year (5% vs. 15% chance).

    More importantly, this is a classic case of looking at raw figures and trying to draw conclusions from them. In this case, the implication is that welfare causes women to be three times as likely to have had a child recently, and that's clearly not the case.

    For instance, what else do we know about women on welfare, compared with women not on welfare? They may be poorer, less educated, have worse job prospects. They may already have at least one child, which increase the likelihood that they're having children. And so on.

    Just consider that the study shows, plain as day, that women were much more likely to have had a child in the last year if they were poor, whether or not they were on welfare!

    If you equalize all other factors, welfare may turn out to have no effect at all. Or if you simply look at the effect of granting women welfare, or compare women in states with different incentive structures, you may find the same thing.

    Finally, this was just an example of someone cherry-picking the one result which looked (and I stressed looked) damning. For instance, he could instead have said women with graduate or professional degrees were much more likely to have had a child in the last year than women with less education. Yet that would be a misleading statistic, too, and for the same reasons.


  9. bobbo says:

    James–I'm glad you see the deficiencies in the census data so clearly. I don't. Some kind of analogy or explanation using totally different words needs to be formulated, then maybe I could see what you see.

    Regardless, we might have "fun" making a list of all those things that people in fact do want to do who are not motivated to do so by receiving money for it? What else can we add to that list besides having kiddies?

    Lets take murder for hire. I think many people are motivated from time to time to kill for money BUT most people don't because while the desire for money is there, the fear/moral values against killing is also strong. But people do kill for money. I assume the percentage in the general population is so small, academic studies may not turn it up==very dependent on the targeted study group? Still–I'm quite willing to say that "in general" people are not motivated to kill for money although "some" are. Of those people who kill who are not in the heat of the moment, the percentage may be significant?

    Likewise with single women having kiddies. May be too small to measure compared to general population but significant among welfare recipients?

    I don't know, I'm just playing with the structure of the studies and trying analogies on for size as I don't see anything directly.

    Thanks for the compliment. Even when I scowl, my motives are pure.


  10. James says:

    I wouldn't describe these as deficiencies in the census data, bobbo. Maybe that's where you're having trouble understanding what I'm saying.

    You can't simply compare women on public assistance to those not on public assistance, and say that any differences between them are the result of public assistance.

    For instance, women on welfare are more likely to be poor and uneducated. So let's compare women on welfare with other women who are poor and uneducated. Lo and behold, these groups look remarkably similar in their childbearing, suggesting that welfare incentives aren't making a difference.

    May be too small to measure compared to general population but significant among welfare recipients?

    I'll grant you that there could be a tiny percentage of women on welfare who have children for the additional benefits. Perhaps they even mistakenly believe that the benefits will more than cover the costs of raising the child.

    These women could be missed by a research study. However, if they make up such a small percentage of the welfare population, there simply isn't a significant problem with welfare creating perverse incentives to have children. That's the issue here.

    If these women were numerous enough to be significant among all welfare recipients, however, they should show up in the kinds of studies that have been conducted.


  11. bobbo says:

    James–note I'm not disagreeing with you, I'm just not understanding what you see clearly.

    I thought I saw the light, and then it got dim again.

    You say: "Compare women on welfare to >>> women not on welfare who are equally poor and uneducated and their birth rates are the same." /// Can you be equally poor and NOT on welfare? Just to twist that a bit, seems like the more educated poor women are ON welfare? ((Educated being defined as knowing about welfare benefits?))

    "Something" is missing in my skill set in order to understand just exactly what is being measured/assessed.

    For me, the situation clearly calls for well founded agnosticism!!


  12. James says:

    Can you be equally poor and NOT on welfare?

    Yes, you can, since income isn't the only qualification to be on welfare.

    You could also compare women on welfare to those with incomes slightly higher, and see if there's much of a difference. The way this is done professionally is to determine statistically how income affects the outcome, having children, and factor that out, so you can observe the independent effect of being on welfare.

    For me, the situation clearly calls for well founded agnosticism!!

    I hear you, bobbo, and I respect your reluctance to believe anything that you aren't trained to evaluate yourself.

    For me, there have been sound studies done which suggest one answer … and nothing to suggest the other possibility. So I think there's good reason to believe that welfare doesn't affect child bearing. Certainly those who claim that it does have no basis for that belief, unless there's more information that you and I aren't aware of.


  13. bobbo says:

    Well, lets not go overboard just because I pulled away from my own experience as a basis for rational public policy?

    My general experience/belief that money motivates people to do things they already want to do has not been "disproven" and still forms the basis for my personal "leanings."

    From Memory–Your studies rely on raw numbers without the more personal individual family histories/interviews that my father undertook. Isn't that a "defect" or want in those studies? Asking the surveyed group "Why are you on welfare?" and others more subtle about their child bearing choices is needed before I would accede the point.

    Money talks while surveys walk. If the position is correct, why give welfare mom's any money at all? They are given money to "help" with raising the kids but then we are supposed to turn that on its head that the "help" will not lead to more of the same behavior.

    A good friend and I are going to make sausages tomorrow. He is somewhat of a statistics expert. I'll put this issue to him and report back (on the issue, the sausage will be great!)

    Don't award that prize to yourself just yet!


  14. bobbo says:

    PS–Back to the Census study showing mom's on welfare have 3 times the fecundity. YOU say if you control for education and income that the birth rate is the same. (How do you know that–just applying your originally referenced studies?? Would that be valid?)

    Well, seems to me its self definitional: if you have a kid and go on welfare to some degree you could be having the kid to get the money OR atleast be more willing to have the kid because you know the welfare money will "help."===how can you really tell without asking the moms? Conversely, the other self selection is going on also. Mom's who are poor and uneducated who do NOT go on welfare obviously did not have the kiddies to get the welfare support. I don't think you have to ask them–but maybe you should as they (their kids) would benefit from the support?

    So–I think your dismissal of the census data is just a bit too easy and suspect on its face. I have to wonder if the survey group of poor uneducated (single?) women not on welfare is statistically significant? What was that number in any given year????

    I love all those jokes that end with "You never asked me." Surveys that don't do personal histories are defective and don't actually answer the question being asked. Haven't we agreed in the past that such studies only show correlations? No proof at all? That has to be true even when the correlations support our positions, No?


  15. James says:

    My general experience/belief that money motivates people to do things they already want to do has not been “disproven” and still forms the basis for my personal “leanings.”

    You can believe what you want, bobbo. I'd be hesitant, though, about elevating anecdotal experience or my instincts about common sense over well-designed scientific research.

    No one doubts, for instance, that money can motivate people. Ask yourself, though: would a woman have a child for $15. For $50? How much would it take? How much would she expect to make for having that child, after the cost of raising a child is subtracted? Just how much do you think welfare benefits exceed the costs involved?

    Thus I think there's nothing obvious about how money-as-motivation affects whether welfare causes people to have children. As a result, the fact that the research all points in one direction is probably our most powerful clue to the answer.

    Your studies rely on raw numbers without the more personal individual family histories/interviews that my father undertook. Isn’t that a “defect” or want in those studies?

    I am not an advocate of statistical analysis over more qualitative, soak-and-poke research methodologies.

    However, I think we need to consider what the question is. If the issue is whether or not welfare generates incentives to have additional children, and if so, to what extent, then we're firmly in the terrain best navigated by quantitative analysis.

    If we know from the "raw numbers" that welfare does not cause women to have more children, what is the defect and how do personal interviews help? We want to know whether or not women on welfare will, in fact, have more children because they are on welfare–not whether they think they do, or will tell case workers that they do.

    Of course, if we're asking a different question (such as why we observe what we do), then a methodology such as interviewing subjects (in a carefully neutral way) might be quite appropriate.

    Perhaps the problem is that the blog you found was citing raw numbers, and now you're speaking as if the studies I cited used raw numbers. In fact, scientific studies only use raw numbers to draw appropriate conclusions; for other analysis, researchers will, for instance, perform regression analysis to remove the effect of confounding factors. Consider the issue above, where we noted that poor women are more likely to be bearing children, so women on welfare can't be directly compared to women with more money.

    They are given money to “help” with raising the kids but then we are supposed to turn that on its head that the “help” will not lead to more of the same behavior.

    I'm not sure I see the problem here, bobbo. We give them money to help with raising a family, and try to design the aid so that it will not encourage them to have more children than they otherwise would. Isn't the right?


  16. James says:

    Back to the Census study showing mom’s on welfare have 3 times the fecundity.

    Again, three times the likelihood of having given birth in the last year, compared with all women in the population.

    Women on welfare are inevitably poor, and poor women are also much more likely to have recently given birth than all women are.

    YOU say if you control for education and income that the birth rate is the same.

    I said that the rates are similar. I implied that if you factored in everything, you'd see minimal or no difference. That's a fair assumption, given that research consistently shows that there is, in fact, no difference caused by being on welfare.

    if you have a kid and go on welfare to some degree you could be having the kid to get the money OR atleast be more willing to have the kid because you know the welfare money will “help.”

    You're making an assumption, bobbo. Let's say a woman without much income inadvertently gets pregnant. She goes on welfare, but in no way did she have the child for the (mythical) money, or necessarily was more willing to have the child because of the aid. She might have been less likely to have an abortion or abandon her child, because the government would help her make ends meet; of course, this isn't necessarily contrary to public policy.

    how can you really tell without asking the moms?

    In general, social science holds that asking people can be a useful tool of analysis in the right circumstances, but can also be a lousy means of getting at the truth, and must always be used carefully.

    Do the women know why they did what they did? Can they accurately state all factors in their decision making processes, and weigh their relative importance? Will they do so honestly for a stranger? Especially one they may see as having power over their benefits?

    Mom’s who are poor and uneducated who do NOT go on welfare obviously did not have the kiddies to get the welfare support.

    That's not true at all. There are several qualifications to go on welfare, and many reasons why not all who qualify will do so.

    I think your dismissal of the census data is just a bit too easy and suspect on its face.

    You're asking some great questions, bobbo.

    However, I dismissed the census data because it was plainly inadequate to answer the question being discussed. None of the issues you're raising, right or wrong, changes that. That data simply can't be used to conclude what that blogger was implying it could.

    I have to wonder if the survey group of poor uneducated (single?) women not on welfare is statistically significant?

    It was. More importantly, there were huge numbers of women in each of the major income groups, and there was a clear trend towards more childbearing at the lower income levels, demonstrating conclusively that it would be impossible to assume that income level has no effect on child bearing rates.

    Surveys that don’t do personal histories are defective and don’t actually answer the question being asked. Haven’t we agreed in the past that such studies only show correlations?

    It's true that causality can't be proven in that way. However, if being on welfare and having more children are (or aren't) correlated, after factoring out everything else that seems relevant, we do know something. In particular, if the children are coming after going on welfare, we know that there's either a factor we haven't thought was relevant, or else welfare does (or doesn't, depending on what we see) have the predicted effect.

    On the other hand, doing personal histories may tell us remarkably little. Do we conduct enough to be statistically significant for each subgroup? And, just as statistical association without more only tells us correlation, all we know is what the subjects tell interviewers. They may not know themselves, and they may not be being candid.


  17. bobbo says:

    James–no notice received yet as to your response. I just checked in to say my friend did not show. Busy guy.

    James–hah, hah. NOTHING more arrogant than to say that social scientists can "prove" human motivation by looking at raw numbers yet NOT LOOK to the people themselves. Thats not YOU being arrogant–that is the surveyors. You say that such interviews are useful only if they are carefully done. This suggests these surveyors cannot carefully conduct a study, yet you rely on their faulty/non-existent logic anyway. ((Only half joking here, but still half joking.))

    Once we admit that these studies only show correlations, isn't any conclusion simply wishful thinking?

    I also think that electing or avoiding abortions is a big wild card here. All the mothers here elected to give birth, if they had an abortion alternate. You even admit that "She might have been less likely to have an abortion or abandon her child,"–that might is not resolved/analysed or evaluated in these surveys.

    But lets forget all the foregoing. I am mostly reacting to my own individual bias: I have been told 436 times in my life that "money is not the issue here" and in each of those 436 times, money was exactly the issue. People lie, cheat, steal and fool themselves about money all the time. Your suggestion that welfare mom's who get money if they have a kid are not motivated to have the kid by money is number 437. But you would have to interview me to find that out, and I may not know what my own motivation is.


  18. James says:

    NOTHING more arrogant than to say that social scientists can “prove” human motivation by looking at raw numbers yet NOT LOOK to the people themselves.

    Not human motivation, bobbo. Human behavior.

    If researchers can demonstrate that people do not, in fact, have more children when they face certain welfare incentives, then it's likely that those incentives do not, in fact, motivate people to have more children.

    Notice how I'm using the term "motivate" in that sentence. This may be a subtle point, but it's important. Another way to think of this may be to observe that the researchers are looking at the people themselves; it's just that to say how people behave under certain conditions (rather than, say, the thought processes at work), what's important is what people do.

    Once we admit that these studies only show correlations, isn’t any conclusion simply wishful thinking?

    Not at all. Look carefully at my previous comment. I noted, perhaps too subtly, that if we know the relationship between being on welfare and having children (the two variables of interest), and we feel we've taken any other relevant variables into account (income level, education, whatever), then the timing of the two variables is vital. In other words, if there's no other variable causing the two variables we're interested in, and one variable occurs earlier than the other in time, we can reasonably believe that we've nailed down causality.

    I also think that electing or avoiding abortions is a big wild card here.

    How so, bobbo? The issue is whether or not women on welfare are likely to have more children, not how they manage to avoid having children. The whole topic of whether they have sex, whether they use birth control and do so effectively, and whether they have abortions or not, would be another issue, wouldn't it? If not, how should it have been taken into account in the studies we're discussing, in order to answer the question they (and we) were interested in?

    I have been told 436 times in my life that “money is not the issue here” and in each of those 436 times, money was exactly the issue.

    That's fine, and it's no doubt true.

    But there are situations in which money isn't an issue. (I'll remind you that I've noted repeatedly that you're assuming there's even a substantial amount of money to be "made" from having an extra child while on welfare. I very much doubt it.)

    If, in fact, people in a given situation do not alter their behavior for money, then money isn't the issue. Whether or not that's true is an empirical issue, and your sense that people often deceive themselves about whether money matters is beside the point, in such a case.


  19. bobbo says:

    James==you are a trip. Few people can argue as steadfastly as you do without getting frazzled and extreme which you avoid completely. You are a perfect blog moderator. Too bad there isn't more money in it heh?

    Well, how subtle can I be? You are right, I wasn't focusing on motivation vs behavior and I did equate them. Aren't people who are motivated to do something more likely to do that something compared to people who aren't motivated to do that something? If studies show that not demonstrated, is that evidence that motivation does not affect outcome, or that the study is flawed? How does subtlety or lack thereof explain this?

    Right now, I'm thinking our disagreement actually focuses on what the concept of "controlled for" means. Maybe the non-subtle meaning of correlations versus causation as well?

    Controlled for–how does one control for income? Most welfare Moms are not under-employed or low paid moms, they are NOT WORKING at all as such payments reduce the total payment making not working at all the more rational choice. What you said was: "More importantly, there were huge numbers of women in each of the major income groups, and there was a clear trend towards more childbearing at the lower income levels, demonstrating conclusively that it would be impossible to assume that income level has no effect on child bearing rates." Well==just the opposite. According to you, the studies say "having money" is a disincentive to having kids. I agree with that. There is a disconnect in the life/data stream isn't there? Having NO MONEY and only one way to get money is not on the same curve as HAVING less and less money? Is this just semantics, too subtle for me, or is this revelatory of the studies flaw?

    Well lets turn to causation, again. Raw statistics can validly answer a few questions like: Do women on welfare have more or less kiddies than similarly aged/educated/situated single women who are not on welfare? I accept the study says "No Difference." From this does it flow that receiving welfare does or does not motivate a significant number of women to have kiddies?

    You say it does, I say it does not. Subtlety is like a box of chocolates. Which one of us got the mint?


  20. James says:

    Few people can argue as steadfastly as you do without getting frazzled and extreme which you avoid completely.

    You're right–patience is a necessity in this game, isn't it?

    Aren’t people who are motivated to do something more likely to do that something compared to people who aren’t motivated to do that something?

    But that's precisely the issue, bobbo. Are people motivated to have more children when they're on welfare? We don't know that; you're suggesting it might be true.

    We don't know that welfare recipients earn enough extra money raising a child that they're tempted to have a child in part for that reason. We wouldn't know that by asking what they think they're doing, either. We know that by looking at what they do, and welfare recipients don't have more children when welfare rules mean additional payments for additional children.

    I used the term "subtle" to refer to two slightly different uses of the word "motivate." If welfare recipients don't have more children, then they "aren't motivated" to have more children … even if they may have a motivation to do so. They simply aren't acting on that hypothetical motivation. Likewise, they may have a motivation to do something, but don't do it, for other reasons.

    Controlled for–how does one control for income?

    With the techniques of regression analysis.

    Which can include zero income, or can include public assistance as income, or be calculated both ways.

    Having NO MONEY and only one way to get money is not on the same curve as HAVING less and less money? Is this just semantics, too subtle for me, or is this revelatory of the studies flaw?

    I'm not sure what you're talking about here, bobbo. You seem to be assuming, again, that having children allows welfare recipients to earn money. In fact, the additional payment for a child is probably swamped by the costs of raising a child; I've seen nothing to suggest that after those expenses, there's a net profit, much less one large enough to motivate one to have a child.

    In any case, I think you're missing the point about income. If welfare mothers have children at roughly the same rate as other, relatively poor women, then they don't seem to be caused to have more children by welfare incentives. The fact that they have more children than wealthier women becomes irrelevant.

    This doesn't mean that welfare recipients might not be so poor that making some money by having children, if that's even possible, would be tempting. It means that they don't seem to be having more children than those in similar socioeconomic circumstances but not on welfare, so they seem not to be having more children at all.

    Now, it's true that poor women not on welfare are mostly (but by no means entirely) of slightly higher income than those on welfare. This is handled by mathematically in one of several ways, for instance by projecting the effect of low income to a lower income status, or by looking only at the population of women who have low enough income to qualify for welfare on that basis, but who don't have welfare for some other reason. (We're also simplifying by talking here about income, when we might really look at income, financial support, material resources, etc.)

    From this does it flow that receiving welfare does or does not motivate a significant number of women to have kiddies? You say it does, I say it does not.

    Why, bobbo? If, taking into account age, education, etc., women on welfare have the same number of children as those not on welfare, why wouldn't you say that they aren't motivated by welfare incentives to have more children?

    If you're saying that welfare incentives don't cause them to have more children, but "motivates" them to do so in the sense of encouraging them to, then that's an interesting issue. What does it mean to provide a motivation, if it doesn't cause anyone to act differently? Many people would have other reasons not to have children, of course, but wouldn't a few more people decide to?

    If you're saying that the research seems to suggest welfare incentives don't cause women to have more children, but doesn't absolutely prove it, that's certainly correct. We can't have absolute certainty here. But it would make no sense to believe that welfare probably does have this effect, if the available evidence indicates the opposite.


  21. bobbo says:

    James==how can we address this issue with a fresh approach? We are precipitously close to just repeating our positions until one of us gets tired?

    I believe the study is "accurate" in its survey results. Its the conclusions drawn that you equivocate on that I disagree with. ((Yes, you equivocate between the study shows welfare does not motivate and that the study is inconclusive.))

    I owe it to this discussion to take another look at that study.

    Well here at para 6 of the main link is an interesting factoid: "Unfortunately, return rates were also high, with 45 percent of ex-recipients returning to AFDC within 1 year. Persons who were likely to use AFDC longer than the average time had less than 12 years of education, no recent work experience, were never married, had a child below age 3 or had three or more children,…"

    When I say that welfare motivates women to have children the above fact from your study is my proof. It matters little that the above statistic can be diluted by a larger study. To somehow lose the above fact is about as valid as the comment in para 7 that "women don't receive welfare, children do." Uh, huh.


  22. James says:

    you equivocate between the study shows welfare does not motivate and that the study is inconclusive.

    Bobbo, what exactly did I say the study was inconclusive about? I don't believe I've ever said that any of these studies was inconclusive in terms of their main findings; like any studies, they can't be the definitive word on the subject, but their conclusions seem quite sound to me.

    I say that welfare motivates women to have children the above fact from your study is my proof.

    Bobbo, what do you mean? That line from the paper said that having children was one of several risk factors for being on public assistance longer than average. This has nothing whatsoever to do with whether welfare causes people to have more children, does it?


  23. bobbo says:

    HEY!!==quit making me eat crow. When I made that comment, I knew you would challenge me so I went back and checked to see when you were "positive" that welfare did not motivate moms to have kids. I didn't think to confirm the equivication that you did not as it was "in my mind" from having just read your post #20. But returning to that post, I see what I reacted to was your use of the word/concept/semantics of "seems" which I avoid like the plague as being equivocating or poor usage. But the entire paragraph in full glory only shows poor usage/style and not equivocation.

    Might this be our difference: I keep saying welfare encourages women to have kiddies and you keep saying welfare does not encourage women to have MORE kiddies. Throw in abortion and both our positions are consistent with one another.

    Hypo==you ask the mom's on welfare if they would have had the kid without welfare and 80% answer "no." Would that be any evidence of the motivational impact of welfare payments or only evidence the recipients don't know their own minds?


  24. James says:

    Might this be our difference:

    No, sorry. :-)

    I think I'm simply trying to be precise, by saying that welfare does/doesn't encourage women to have more children … than they otherwise would. Certainly the literature isn't making that distinction.

    Now, it's possible that your belief somehow involves abortion, in a way that wasn't specified. That I couldn't say.

    My instinct, though, is that you're convinced that welfare must, somehow, cause women to be more likely to have children; you seem reluctant to acknowledge the evidence that your instinct isn't born out, and even that there aren't financial incentives in the welfare system to encourage that behavior in the first place.

    Would that be any evidence of the motivational impact of welfare payments or only evidence the recipients don’t know their own minds?

    That depends, doesn't it? If the facts suggest that mothers on welfare don't have children they wouldn't otherwise have (or that only 20% of them do), then those recipients obviously don't know their own motivations (or aren't being honest with you).

    If the facts suggest that something like 80% would be having the child because of welfare, then it seems you can trust the answers you're getting. Of course, then you already knew what they're telling you, from a much more reliable source. In this case, who has a child is probably a much more reliable indicator than what they say about it.


  25. bobbo says:

    Heh, heh. I like to think I'm open to changing my mind, but on reflection the evidence is I don't often do it. I've always thought it was because on so many issues, I don't actually carry a portfolio.

    Do I really think that welfare payments motivate women to have kids they otherwise would not? = Yes.

    Should this result in welfare mom's having more kids than controlled for mom's not on welfare? = Yes.

    Does it? = No.

    I'm being rational so far. So, what to do when objective studies clash with long held beliefs? I can see welfare motivating moms to have one kid and the experience is so time consuming/interfering in life that even welfare moms don't think the money is worth it so they stop. Welfare payments would still have motivated an increase in births without any difference in the two groups because of the variable of abortion.

    So. Extraordinary claims require extraordinary proof. In my view, claiming money does not motivate (and achieve) its spent on purpose is extraordinary in my view, and I don't see studies that don't even ask the welfare moms why they had kids as being extraordinary.

    I will however back off one notch on my conviction that welfare payments result in more kids AND be open to receiving more information supporting the counter intuitive position.

    Don't ask me how many notches I'm working from.


  26. James says:

    claiming money does not motivate (and achieve) its spent on purpose is extraordinary

    Well, bobbo, please bear in mind that you're the one suggesting that there is money involved to motivate welfare recipients. I've suggested repeatedly that welfare payments for a child would have to be set at a level much too high in order for women to turn a sufficient profit, after the costs of raising a child, to consider it worthwhile to endure pregnancy and labor, and to spend years raising a child.

    I've also pointed out that I don't think it's fair to say simply that money motivates people to do things. Would $5 motivate you to kill someone? No. The money has to be enough to overcome any reluctance you have, any risks you have to bear, and so on.

    How much money would it take for you to have a child you otherwise wouldn't? What about the risk that the welfare payments would stop in the next eighteen years or so? What about the time and effort of raising that child? Would you do it for $5? For $100? $500? Remember, you're raising that child, which is enormously time-consuming. Just how much do you think these welfare payments are set, over and above the material costs of raising a child?

    I appreciate that you find it counter-intuitive that welfare mothers aren't having children for the money, but this just doesn't square with my own intuition.

    I don’t see studies that don’t even ask the welfare moms why they had kids as being extraordinary.

    I don't believe that interviewing welfare mothers, and asking them whether they think they had a child for additional welfare payments, would add nothing to a sound statistical analysis of large numbers of welfare recipients and whether or not they actually have children. And I'm afraid that I'm still not sure why you believe otherwise.


  27. bobbo says:

    Well, bobbo, please bear in mind that you’re the one suggesting that there is money involved to motivate welfare recipients. /// Who created this thread? But I'll take the honor as I don't see who came first has any relevance to the answer. Also==there is money involved. Thirdly, I thought we agreed it was a motivation and we are just disagreeing on whether or not it is sufficient to cause mom's to have kids to get that money? Must be that subtlety issue again?

    I’ve suggested repeatedly that welfare payments for a child would have to be set at a level much too high in order for women to turn a sufficient profit, after the costs of raising a child, to consider it worthwhile to endure pregnancy and labor, and to spend years raising a child. /// And I have responded each time that women who don't want kids will abort while those who do want kids will have them anyway and those on the fence should be motivated to fall towards the money, at least once.

    I’ve also pointed out that I don’t think it’s fair to say simply that money motivates people to do things. Would $5 motivate you to kill someone? No. The money has to be enough to overcome any reluctance you have, any risks you have to bear, and so on. /// I did try to make the point that women who don't want kids have abortions. Its women that are motivated anyway, or on the fence (yes a finer shaving), that would be motivated further to have kids. I do buy CRAP all the time that I have a casual interest in but don't really want until I walk into a store and find a two for one sale or other financial incentive. Money did not convince me, but it was another factor.

    How many welfare homes have you been in? I don't think "I" am the typical welfare mom. Young, uneducated, jobless, single and pregnant. Is this a group well regarded for their sound decision making? Most/many/some/a few welfare kiddies are "raised" with great neglect.

    I appreciate that you find it counter-intuitive that welfare mothers aren’t having children for the money, but this just doesn’t square with my own intuition./// Intuition yes but also listening to my dad who was an expert in the field and my own experience as a slum lord. Anecdotal but not just whimsy.

    I don’t believe that interviewing welfare mothers……. /// I find that very arrogant. You are familiar with people with arrogant opinions yes? Didn't you say they would only be valid if carefully conducted? But if YOUR mind is already made up, why carefully conduct anything?

    a sound statistical analysis of large numbers of welfare recipients and whether or not they actually have children. And I’m afraid that I’m still not sure why you believe otherwise. /// Well, WAIT for the next sausage day and maybe my expert will inform us both? But I'll say it a third time: just because welfare mom's and non-welfare moms both have the same number of kiddies doesn't mean they were motivated the same. Statistics don't go to the soul of a person. I still don't know how you can say you understand correlations are not proof and then act as if they are.


  28. James says:

    Who created this thread?

    Bobbo, I raised the issue of welfare encouraging childbearing in order to debunk it as a myth. You, on the other hand, keep insisting that you don't believe the empirical evidence on this count, because you're sure that money influences decisions. My point is that your argument here depends critically on believing that welfare recipients can make substantial money bearing and raising children, and I doubt that's true.

    Yes, there is "money involved," but as I've said countless times by now, I have no reason to believe that welfare payments are mistakenly set so high that people net a significant profit after the expense of raising a child, enough to even begin to compensate for the time and trouble.

    those on the fence should be motivated to fall towards the money

    If there is a profit to be made, and if that money outweighs the downsides to having an otherwise unwanted child for some people, then maybe some of those people will behave as rational economic actors and have children they otherwise wouldn't. The other possibility is that people decide to have children largely for psychological or sociological reasons, and aren't influenced by considerations of profiting from their children.

    That's a lot of "ifs" in my book, and on the other side of the ledger, there's the fact that the evidence shows that there don't seem to be people ending up on the other side of the fence.

    I did try to make the point that women who don’t want kids have abortions.

    I don't see where abortion enters into this. My arguments are entirely independent of whether women decide not to have children by not having sex, by using contraception, by using it more carefully, or by aborting any unwanted pregnancies.

    Is this a group well regarded for their sound decision making?

    Hmm. That seems to me, bobbo, to argue for looking at how this group actually behaves in response to incentives, and not to simply assume that they must respond rationally, in carefully calculated ways, to economic incentives.

    Maybe these young, uneducated people are having, or not having, children for their own reasons, and paying little or no attention to the details of how the welfare regulations would change their payments if their life circumstances changed? It's a possibility.

    Didn’t you say they would only be valid if carefully conducted? But if YOUR mind is already made up, why carefully conduct anything?

    Bobbo, I simply said that interviews wouldn't contribute anything as a methodology to answer this particular empirical question.

    This isn't arrogance about whether the interviews are well done or not, or about the outcome of the research. It may be arrogance or it may be sound training in social science methodology, but either way, it's a judgment about the nature of research design.

    just because welfare mom’s and non-welfare moms both have the same number of kiddies doesn’t mean they were motivated the same

    I see. So you believe that there might be some countervailing motivation for those not receiving public assistance to have children, which is absent from those on welfare? And that this motivation exactly balances out the motivation derived from welfare incentives, so that the two groups end up having children at exactly the same rates?

    That strikes me as entirely logical, and even possible, but highly unlikely.

    I still don’t know how you can say you understand correlations are not proof and then act as if they are.

    I must not have been clear, then. I tried to explain how we can get from correlation to causation. This is a lengthy subject, but it's a perfectly valid one. In short, correlation doesn't imply causation, but that doesn't mean research designs can't demonstrate causation. They just can't jump to causation from the mere fact of correlation.


  29. bobbo says:

    Good Morning. Do you really want to spend Wednesday on this? Might as well, you are sharpening my "skills." Practice, practice, practice.

    Actually, I'm the one equivocating. I actually do prefer the science of objective studies to "personal opinion." An immediate dodge would be to simply assume and say that the welfare programs in California in the early 1980's has obviously changed the motivations and the outcomes as are the reality today?

    The open contentions then seem like quibbles born of ignorance.


  30. James says:

    An immediate dodge would be to simply assume and say that the welfare programs in California in the early 1980’s has obviously changed the motivations and the outcomes as are the reality today?

    Let's run with that, then, bobbo, and enjoy Wednesday.


  31. Anna says:

    Statistics show the overwhelming majority of unwed mothers are amongst black and hispanics! It doesnt take a scientist or a brilliant person to know to uphold abstinance. YOU wanna have sex, have at it..go for it, but at least protect yourselves so that we, the middle class working Americans of this county, the back bone of society can stop paying for your screw ups! If we were to stop checks coming in to these people, it would be amazing on how the percentage of unwed mothers would drop!


  32. James says:

    Anna, first of all, you have your facts wrong–indeed, backward. In the latest statistics I've seen, for example, a solid majority (60%) of out-of-wedlock births in the U.S. are to white mothers.

    Second, the issue isn't race, it's poverty. Poverty has become strongly associated in this country with single-parenting and, of course, with welfare. Poverty, in turn, is associated with race and ethnicity. It would turn logic on its head to acknowledge that blacks and Hispanics are disproportionately poor in this country, because of both historic and contemporary discrimination, and then to complain that both groups disproportionately exhibit the symptoms of poverty–the same symptoms which poor whites tend to exhibit.

    Third, you can't simply equate marital status with unwanted children, or either of these with receiving welfare. Just look at the research I presented above, showing how these connections don't hold, and how there is no clear relationship, as you would like to believe, between welfare checks and having more children.


  33. Eric Sheptock says:

    I know of families in Washington, DC where at least 3 generations have been or are presently homeless. Google the story of Relisha Rudd, a missing 8-year old homeless girl.


  34. Eric Sheptock says:

    I actually stopped reading about 5 comments back. I was looking for evidence of someone's claim that welfare programs used to forbid people to work. My Google search brought up your blog which looked like it might tell me one way or the other. I don't have time to read any more of your interesting exchange. So, do you know if what the person told me is true?????


  35. James DeWolf Perry says:

    Eric, I'm not an expert on the history of welfare programs in this country, but I've never read anywhere that welfare programs used to forbid people to work. The debates I'm aware of about welfare over the years focused on claims that welfare didn't require recipients to work, which led some people to believe that welfare recipients were being encouraged not to work, and that welfare sometimes discouraged people from working, through rules that would respond to a modest increase in earnings by reducing benefits even more.

Leave a Reply