The Institute on Assets and Social Policy published a report on Monday based on economic data from the same 2000 families from 1984 to 2007 (hat tip Michael Powell). Its sobering results likely understates the case, since it does not include the post financial crisis period. The study found that the median wealth gap between white families and black families rose from $20,000 in 1984 to $95,000 in 2007. Note that home equity was excluded from the calculation.
Essentially, this underlying trend was that of the rich getting richer:
It is striking that even high income blacks have not been able to translate their earnings into greater wealth accumulation. One has to imagine that at least some of the gap comes from the greater representation of whites in the top 1% of the population, which has seen a great leap in its share of total wealth over the last 20 years. The authors argue that tax policies that favored the well off and credit discrimination each played a role:
The racial wealth gap results from historical and contemporary factors but the disturbing four-fold increase in such a short time reflects public policies, such as tax cuts on investment income and inheritances which benefit the wealthiest, and redistribute wealth and opportunities. Tax deductions for home mortgages, retirement accounts, and college savings all disproportionately benefit higher income families. At the same time, evidence from multiple sources demonstrates the powerful role of persistent discrimination in housing, credit, and labor markets. For example, African-Americans and Hispanics were at least twice as likely to receive high-cost home mortgages as whites with similar incomes. These reckless high-cost loans unnecessarily impeded wealth building in minority communities and triggered the foreclosure crisis that is wiping out the largest source of wealth for minorities.
Yves here. This report is a stark illustration that policies that favor capital over labor will, not surprisingly, widen the wealth gap.
Cue in the trolls who will accuse Yves of class warfare..
Sigh!
Sigh. No, cue in someone who will pre-emptively dismiss anyone who might possibly disagree with his opinion as a “troll”.
Oh, I see he got in first, right on time! :-)
OK I will take the bait. How does the presence of Bill Gates, Warren Buffet, etc bias these numbers? How does this data compare with other data on stagnant wages for the 95% of the population which includes all races?
I looked at the linked report and must say that it did not inspire much confidence.
What’s the libertarian response to this? To the oligarchy and plutarchy that result from flat-tax policies (which we already have)?
Does it make sense that your boss is paid a higher multiple over you than your dad’s boss was over him?
Is it because of some efficiency? If so, why do foreign companies (that more typically actually make things people want) exhibit lower such multiples than US companies?
Is it necessary to motivate people? If so, why do foreign executives bother to keep working? And why did your dad’s boss bother to work?
Why do you want to tax capital at lower rates than labor? Don’t you want to encourage people to work?
What exactly are “capital gains” anyway? Capital gains aren’t just an improvement in some business, although that’s what we usually consider. But, that’s *relative* capital gains. There are “absolute” capital gains, too. Do you know what that is? One of the primary inputs is public debt. Every time the government runs a deficit, it’s adding more fiat currency into circulation. That drives up private asset prices. Lately, a lot of the “gains” come from the absurd public debt we’ve been running up. Since the rich demanded the government lower their taxes nominally and helped create our deficit problems in the first place, they must not be allowed to keep the gains. The government must take back these gains via higher capital gains tax from those same primary beneficiaries. (I would’ve preferred balanced budgets and “no bailouts”. But, those ships have sailed.) Any other choice is another victory for the rich in their class war against the bottom 85%
I’ll tackle one part…
Capital gains should be taxed at a lower level than wages because they are earned over long periods of time and are taxed on a nominal basis. That’s also why short-term gains are taxed as wages– they are earned over a short period of time and are not nearly as impacted by inflation as long-term gains.
I would definitely be in favor of taxing capital gains at wage rates, on a real (i.e. inflation adjusted) basis. But I imagine there would have to be a consensus on the correct CPI measure for that to happen (as if that’s possible…)
As with most things, it is more complicated than you suggest. Investments that grow without current income and that have long time horizons actually have the tax benefit of deferral. You are “investing” with the “government’s share of the money” in the second year and each following year. This is the magic of all non-recognition provisions such as like-kind exchanges and 401k’s. I’m not opposed to indexing on interest rates, but the tax would also have to be altered to capture the deferral aspect.
interesting thoughts.
one guy offering a somewhat different view to the idea that tax subsidies for capital are terrible for labour is austan goolsbee – one of pres obama’s economic advisors.
if you like, read the exec summary on the following report:
(warning- it may well shatter your world views.)
“Investment Subsidies and Wages in Capital Goods Industries: To the Workers Go the Spoils?”
http://www.nber.org/papers/w6526
Yves is not guilty of class warfare, of course. But, it would be helpful to understand the increase/decrease in rate of spend on social welfare programs during this period. Yves’ comment suggests Congress stood on the street corner handing cash packets to Daddy Warbucks when in fact Congress was spending like (begging your pardon, USN) drunken sailors.
The closest you got was with the drunken sailors part. If you care to follow the government money over the past 40+ years I think you will see it has gone to expand US imperialism all over the world covered by our military which I am sure you would characterize as just more social welfare.
Please come back when you can tell us how much per barrel of oil we subsidize the distribution and protection of….for starters.
There might be another, simpler explanation for the results of this small sample study:
The Institute on Assets and Social Policy has a mission statement – in bold letters on the first page of their website: “Informing policies and practices that broaden wealth and reduce inequality….”
Do you think that ANY Institute that has a mission might design a study in which they pre-select their sample study in a manner that will be most likely to “prove” that said Institute’s works (and grant money receipts) are “vitally important”?
Be it Right or be it Left in its philosophy and policies, every institute can and will create “data” that “proves” it’s philosophy.
I don’t have time at the moment to review the study in depth, but are the high income blacks and whites initially chosen the same in position, age, education, location, etc? I’ll bet I could create a study that would produce the exact opposite results if I selected young black high earners in say, Silicon Valley start-ups, and older white high earners in say, Detroit auto plants.
Just sayin’….
If they were able to pick a sample in 1984 that would prove their conclusions in 2007 they are very good.
It really wouldn’t be that hard. In fact, if you’re up for it, I’m willing to do a straight wager with you – we both put our stakes in escrow, and I’ll recreate the same study, and I will bet that I can have my black sample outperform my white sample in 23 years. Name your stakes. Game on.
Of course, I will be dead in 23 years, but I will direct the escrow to donate my winnings to Doctors Without Borders.
Care to play?
Sorry Mortimer, not going to be Randolf.
http://www.youtube.com/watch?v=_gekaEzqj5g
Part of the extremely involved methodology of a cohort study is the pursuit of a random sample, usually by taking random samples of random samples. You start out with a sample of 200,000 to get a final sample size of 2,000.
Your team has had since 1984 to prove this wasn’t a random sample.
Actually, it’s on those who produce the study to prove that the sample was truly random, large enough to be representative, etc. They are the ones with the burden of proof (to reject the null hypothesis).
How do you prove randomness? You can’t. You use statistical guidelines in developing the survey cohort. Given the longevity of this study I am going to side with the study authors.
You can prove non-randomness, especially over time. Make a go at it.
bob – LOL, thanks for the Trading Places link.
But for the record, I’m not on any team. And I noticed that you are unwilling to actually take my bet that, without my affecting the lives of anyone (unlike Mortimer and Randolf), I can produce the opposite results of this study. And I even offered my winnings to charity! Seems that you’re not confident in your assertions – either that or you pass up free money.
But definitely, thanks for the link – great film, especially Eddie Murphy’s pickup line to a passing woman while pretending to be a legless beggar – “Baby, once you’ve had a man with no legs, you’ll never go back….”
In case it wasn’t obvious from the link, I find the idea of letting you select a non-random group of people, and then betting on their outcomes in life to be just morbid and reprehensible as the behavior demonstrated in the film I linked to.
The rest of your post boils down to calling me a pussy. I disagree.
Take all of the time you were proposing to put into developing a cohort survey sample (probably thousands of hours) and use just 10% of that to try and find the lack of randomness in the study which is the subject of this post. Science and especially statistics needs skeptics.
bob, As noted below, I can’t find anything that describes how the sample was selected here or that in any other way describes (in detail) the methodology used in this study so that I can even attempt to evaluate it. Where is it? If it’s not there, how can you evaluate it (and decide that it’s legit)? How do you know what guidelines the authors used? Please provide page numbers, etc. TIA.
I can’t find the data either. There is a very good chance that the data, from 1984, isn’t on the internet.
Good long term cohort studies are very valuable. There is a chance that you would have to pay for the original documents, and any data that they include. By now the study is probably well into the thousands, if not hundreds of thousands of pages.
Try writing to the authors.
I still stand by my judgement, if a flaw was going to be found in the randomness of the sample, it would have been found by now.
I should also note that long term cohort studies, by definition, have to include the names and contact information of the people within the survey. Great efforts are taken to protect the identities of the people within the study. Publishing the methodology would have to include scrubbing the data of any personally identifiable information, very costly.
The other option is to make the full study available only to a limited number of people, usually with a few initials after their names.
Finding sample bias in this particular study could lead to a few more initials after their name.
“Good long term cohort studies are very valuable”
But without a good description of the study’s methodology, there is no way to know if this is a good study. I’m not assuming it’s bad, like some of the posters seem to but I’m not assuming it’s good like you seem to be either. I think either assumption is bizarre and completely unsupported by the data that’s available to us.
bob, AFAICS, this study has not been published in any kind of professional journal (esp a peer-reviewed one) so I’m even more at a loss as to why you are assuming that it meets the standards you claim for it.
It’s over 23 years old. It’s a cohort study.
Are you, or have you been on any drugs that have been released since 1984? You, and your doctor are relying on the same survey methodology, over a much shorter time period.
That is a very bad example, they find problems with drug studies all of the time, because they are short term studies.
I am basing my JUDGEMENT on the fact that it is old, and presumably time tested. I do not have access to academic journals from the 1980s, so I can’t refute your point there. I would suggest that the name of the original study, and methodology, are under a different name. Again, write the authors of the paper.
I was a very small part (data collection and aggregation) of a long term cohort study and know the lengths that are taken when setting up one of these studies. They are built to last, this one has.
Since you find betting on the outcome of a study which would not in any way impact the lives of those studied to be “reprehensible”, I suggest you fire of one seriously outraged letter to the people at Brandeis who did this study – because that is EXACTLY what they did – their funding depends on the outcome. They are not only betting on the outcome, they get to stack the deck.
“Hey, if we can prove income inequality between races, our grant money will increase! Let’s do a study! Let’s not make the details of the study available to the public, but let’s broadcast the results that prove how invaluable our work is, as it is needed to correct the problem that our study will be designed to show is RAMPANT.”
I don’t think it would be that difficult. E.g., if you picked whites, especially those with high income, who: were healthier (and whose families were healthier); lived in areas where most people were healthy, had access to good healthcare, water, etc., and had good jobs; were legacies at good schools or in strong unions; had smaller families and richer families (who could leave them more money, pay for their education, etc.); and who had other characteristics highly correlated with the high wealth and blacks who had the opposite characteristics, the data would probably be strongly biased toward this outcome regardless of the true disparity between the races. If I can think of these things off the top of my head, I’m sure that people in the field could be even more inventive. (NB: I’m not in any way suggesting that The Institute on Assets and Social Policy did that or biased the report in other ways. I have ZERO reason to think that this report is biased. However, to think that one could not do that in a longitudinal study seems very naive to me.)
“I have ZERO reason to think this report is biased.”
That’s fair, but since this Institute, like many Institutes, is run by people who are paid by grant money which will only flow in if their mission is perceived by the grantors as vitally important, might you be able to assert that “I have a tiny, tiny suspicion that this report might possibly be biased”?
Of course. Any report might be biased. Most probably are, even if the bias entered unconsciously. ;) Which is why the burden of offering enough proof to reject the null hypothesis is on those who conduct the study. I can’t evaluate this study because the link only provided a brief summary. E.g., I didn’t see anything that described how the sample was selected, how the data was collected, etc. And it’s far from my field so I have no idea whether other studies tend to support or refute this one. (Any study, even if done perfectly and without any bias at all, may come to the wrong conclusion WRT the population one is generalizing to so completely accepting the conclusion of one one or a few studies is never wise IMO.)
The most important “capital” on the planet is scientific and technological know-how. Not those slips of paper. And not even gold. That know-how isn’t something a few people today created. Generations of humans contributed to what we have today. Why do we fashion a society where the benefits of generations of human achievement increasingly flow to a smaller and smaller group?
Ask that question to the people who did it, starting with all of the politicians who voted for TARP. Feel free to ask President Obama first, since he’s been at the helm of perpetuating it since January of 2009.
Because we continue to honor the accumulation of capital/wealth instead of life achievements.
We tried to have a by and for the people sort of thing but it seems to have gotten co-opted. I think we should try again.
Yes, but the leadership of our society is so much more diverse today. These increasing income gaps must be the result of poor individual choices amongst the population.
After all, the new Ms. Trump USA proves that anyone of any color can make it in this great country.
During the time period shown in this graph, unions in the USA have almost gone extinct! Using the advantages of union membership to make your point is, therefore, unconvincing.
The unions, especially trade unions for construction in urban strongholds, have not helped African Americans and Hispanics as well as Irish Catholic, Italian Catholic and other European immigrants who built the unions to begin with. The strategic decision by DuBois, in founding the NAACP, while a necessary organization, was not a sufficient one to garner economic gains commensurate with political ones. Unions are facing a double barrel attack by Black and also, right wing political leaders for not opening up the ranks to greater minority participation. The route of Booker T Washington, which promoted practical, vocation trade skills as a path to empowerment, was and is denigrated by DuBois and his school of thought. Today, the unions alone stand in stark contrast as survival organizations, preserving economic buying power, and some privileges, such as medical benefits and college tuition payment for children. Many working class whites receive the benefits of unionization in the form of middle class standards of living, while Blacks are falling behind, especially those without professional credentials from universities. Also despite over 110 traditionally historic Black colleges, with over 900,000 matriculated Black college males, the middle class is less and less of a prospect, as the job opportunities do not seem match their white counterparts of equal educational attainment. Compared with approx 210 Catholic colleges in the USA, a proportion commensurate with the % of Catholics in the country. The institutional infrastructure for success seems to be in place, but not the logical progress to success.
While unions are not a white privilege, but a working class survival organization, the right wing would like to characterize them as organized crime laden goons and a racist institution that should be opened up to minorities, like a suburban swim club. There are other variables as well as over all discrimination, and poor educational resources before college for the 80% of the population that does not go onto college. The countervailing benefits of unions for this group have largely been lost. But, the general population as a whole is not faring better. The Black population serves more as the canary in the coal mine than as the anomaly that points out only racial discrimination. The work of Robert Perrucci and Earl Wysong, “The New Class Society”, argues that a new economic stratification is clearly emerging, wiping out the hallmark of the great middle class component of American Civilization. Blacks and Hispanics, with absolutely minute participation in the elite credentials of Harvard, Yale and Princeton, along with others not sharing similar membership and jobs in their corporate outposts, Goldman, GE, McKinsey, (sorry Yves, present company excepted)The Pew Trusts, The Ford Foundation, etc will find themselves relegated to permanent menial servitude, and even neo involuntary servitude, as THE NEW PEONAGE, replaces the New Deal. Much of the rest of society will be joining them soon after their home foreclosure and underemployment happens with depressed wages leading to nowhere.
Here’s a libertarian perspective (a la Charles Murray): Blacks have low IQs and would not thrive in a meritocratic society so they deserve their crappy social status and your contempt.
That is an incredible mischaracterization of Charles Murray’s thesis, and your assertion that he or a “Libertarian perspecitve” finds blacks to be contemptible is, well, contemptible.
Try to have an honest debate if you choose to engage in debate.
It may be a mischaracterization of Charles Murray, but at least Murray Rothbard takes that argument:
“Two reasons we have already mentioned; to celebrate the victory of freedom of inquiry and of truth for its own sake; and a bullet through the heart of the egalitarian-socialist project. But there is a third reason as well: as a powerful defense of the results of the free market. If and when we as populists and libertarians abolish the welfare state in all of its aspects, and property rights and the free market shall be triumphant once more, many individuals and groups will predictably not like the end result. In that case, those ethnic and other groups who might be concentrated in lower-income or less prestigious occupations, guided by their socialistic mentors, will predictably raise the cry that free-market capitalism is evil and “discriminatory” and that therefore collectivism is needed to redress the balance. In that case, the intelligence argument will become useful to defend the market economy and the free society from ignorant or self-serving attacks. In short; racialist science is properly not an act of aggression or a cover for oppression of one group over another, but, on the contrary, an operation in defense of private property against assaults by aggressors.”
So according to Murray Rothbard who seems sympathetic to The Bell Curve thesis, blacks are too stupid and libertarians should invoke The Bell Curve argument to defend the economic status quo where blacks have little wealth, low income, and occupy jobs with low prestige to defend the regime of private property. According to Rothbard, if blacks are less intelligent, then their very low economic status is the just outcome of the meritocratic free market and no effort to ameliorate their low social status should be undertaken because that would disturb the just natural order.
I don’t think it’s right to look at racial indicators. Blacks may or may not have lower ability than Whites. What happens, happens. Certainly it’s unreasonable to try any longer to “fix” this “problem” that resists every effort at fixing. Treat each individual fairly — and that’s everyone’s responsibility as a human being. The government with their rules and enforcement cannot do more than they are doing, and they have gone way to far to say we have to treat some more fairly than others to “fix” a “problem”.
In other words, I don’t care about the statistic reported in this story.
I read reports that black neighborhoods were especially targeted by banks for subprime loans. I wonder if that was a factor.
One man said better than all the rest of his ilk, see:
http://www.youtube.com/watch?v=mn4daYJzyls
Skippy…or is the horse’s mouth to random…eh.
Yes, the rich are getting richer and the poor poorer.
I also notice these
1) skinny people getting skinnier and fat people fatter.
2) pretty people getting prettier and ugly people uglier.
3) tall people getting taller and short people shorter.
4) healthy people getting healthier and sick people sicker.
5) dishonest people more dishonest but honest people, well, it’s hard to say.
Surely you understand how very wrong your examples are – on so many levels, don’t you?
Maybe not naturally but by artificial means.
Of course, perhaps I should put, to use just one example, pretty people in quotations marks, like, ‘pretty people,’ to signify by ‘pretty,’ it’s subjective. So, you can have more people having plastic surgery, looking artificial ‘prettiers’ by someone’s subjective standards. Realtively speaking, other people become uglier, in their subjective and artificial sense, of course.
And of course, by fat people, I don’t necessarily mean the same fat people, but I mean the obese people you saw in the ’70s were not as obese as you see today just because our diet habits, or the anorexic people are more extreme now.
As for healthy people getting healthier, because of the increaing toxic environment, people who maintain healthy habits with healhty life styles will be healthier, relative to those who don’t and therefore be sicker, relatively speaking.
By the way, I also see that
6) leftwingers are more left wing and rightwingers more right wing.
You don’t see any of these?
Aki_Izayoi :
So when you are eating your own FECES, do you masturbate at the same time to photos of Ron Paul?
This world is not for you. This web site is certainly not for you!
Does fucking naked capitalism tolerate this type of blatant and unmitigated racism? I guess so because the post is still burning a hole into this site.
Maybe you should move to registered accounts?
In fact, I’m going to come back here from time to time and if see more of this, I will force you to implement registration.
I think you mistake that comment as serious. I beleieve it was snark. There was a previous article up-http://www.nakedcapitalism.com/2010/05/what-is-the-proper-libertarian-response-to-the-deepwater-horizon-oil-spill.html
It was hypothesizing “Libertarian Responses”….which is what I beleive the commenter was riffing off of…..
Yves, I hate to be the one to bring this to your attention, but…
This “Racial Wealth Gap Study” was NOT a study about the wealth gap between whites and people of color. It was actually a study about Confirmation Biases and Selective Perception on the part of authors and bloggers.
Seriously, what I’m writing here is true. The social scientists at Brandeis wanted to see how susceptible commentators are to using studies which provided NO support. None. All they did in this fake study was to provide a simple-minded conclusion designed to conform with pre-conceived worldviews.
Really, Yves. I expected more from you and quite frankly, I’m surprised you fell for it.
[Incidentally, I’m told that was much internal debate other the engineering of the fake study. Some of the architects felt the lack of credibility of the fake study was too obvious and that no thinking person could possibly fall for the ruse. They were wrong.]
I’m sure at this point you are scrambling back to the “study” looking for a link…some bit of support or evidence which supports my disclosure to you. Normally, the authors of the real study would provide support, but since you didn’t need ANYTHING on the first go-round, they reasoned you’d accept the disclosure as fact on the second.
If, however, you are still skeptical. I must say, first and foremost: That’s so refreshing. Honest skepticism. Good for you; you’re really coming along!
Secondly, here are some additional facts to tide you over until the authors of the real study (the study about confirmation biases) deem it appropriate to release their underlying evidence and methodology:
1. You have no idea how many families were used. None.
Could be 2 families or 200,000.
2. You have no idea as to the methodology used in the “study”. None.
a. For example, what role did inheritance play? [Studies have shown it’s a much greater role than anticipated. I know you’re not interested in underlying evidence or anything like that, so I’m just going to let the conclusion stand—on its own.]
b. Were the Ultra-wealthy included in this study? Depending on the mysterious sample size, couldn’t the authors arrive at any conclusion they wanted by merely including 5 ultra wealthy white families?
[You’ve accepted a simple conclusion at face value, simply because you were told to do so.]
3. Finally, and most importantly with respect to the Confirmation Bias Study: You’re a bold-faced liar if you say that you would have accepted a study that ran counter to your pre-conceived worldview with this paltry level of “support”.
—–
BTW: Not that you’re interested in such trivialities, but here’s an actual Conclusion of a study on the same issue, with a critique of the PSID “studies”. (PSID is the grant operation that funded the Racial Wealth Gap Study…the one that Yves blindly accepted…which supports the conclusion of a real study about confirmation biases):
“Using the 1984, 1989 and 1994 wealth supplements of the PSID, this study has examined patterns of wealth
accumulation by race. DURING THE PERIOD, THE RATIO OF AVERAGE WEALTH BETWEEN AFRICAN AMERICANS AND WHITES REMAINED ALMOST CONSTANT. Though the much higher wealth of whites implies that the absolute amount of
wealth accumulation was much greater for them than for African Americans, there were some surprising
similarities in the pattern of wealth accumulation. In particular, (active) savings accounted for slightly more
than half of wealth accumulation for both groups, despite the fact that the rate of savings out of income and
the level of income were both greater for whites. The return on capital and capital gains as a share of the
change in wealth were both somewhat larger for African Americans than for whites. However, inheritances
played a much larger role in wealth accumulation for whites than for African Americans.”
I know it’s antiquated and trite, but I won’t leave you with a mere conclusion. Here’s a link:
http://www.levyinstitute.org/pubs/wp/311.pdf
This last study even has little numbers after a claim. These are called “End Notes.”
Duncan- I love conspiracy interpretations. It’s like looking at a red-colored painting, concluding that the artist secretly hated the color red, and testing this by confirming it with the fact that other viewers see the painting as done well in red shades. Good sleuthing!
That’s what you say! The study was actually designed to provoke furious reactions of spluttering indignation from people who really think growing income inequality is kewl but who are also uneasy over the fact that its “optics” aren’t too hot.
I will probably find these long-winded, semi-serious, sorta-funny salvos to skepticism far more engaging when they attack, rather than support, your pre-existing worldviews.
If median wealth is much higher in one group than in another, even if the the groups have the same income profiles, I would expect that the wealthier group has access to credit on better terms than the poorer group. The old saying is that the bank has plenty of money for people that don’t need it.
Referendum requires filing 76,682 voter signatures by 90 days. But Still at most of online polls people support the Law. make your voice heard on http://immigration.civiltalks.com/
Is it not stupid to fight with Mexicans for a job in the fields at the national level?
Mexicans are hard working, fellow who just came here to do work that Americans, including blacks, will not do!
“The characteristic feature of the loser is to bemoan mankind’s flaws, biases, & irrationality –without exploiting them for fun and profit.”
Nassim Nicholas Taleb, Twitter
From the article:
“African-Americans and Hispanics were at least twice as likely to receive high-cost home mortgages”
*
I get so tired of these vagaries.
Okay, I give up. What exactly constitutes a “high-cost” home mortgage? What is the definition of “high-cost?”
The graph looks disturbing, the high paid wites are probably the top corporate executives, like for example, Goldman Sachs – many of whom are young, good-looking white males. LMAO.
Here’s my take on all this: the world seeks balance. Ebb and flow, up and down, black and white.
This financial crisis is a way of providing balance. Soon, the rich will be poorer and the poor will be richer. Soon, the criminals will go to jail and the wrongs will be righted and the good people will move into positions of power.
Balance. Every 30 years or so, it happens whether we want it to or not. One of those mysteries of life.
Or, not.
“Are you, or have you been on any drugs that have been released since 1984? You, and your doctor are relying on the same survey methodology, over a much shorter time period.”
No we are not, at least not if I’m taking a drug approved by the FDA. ;) That may be where your confusion is coming from. Social science methodology, especially when dealing with a correlational study like this one (as opposed to an experimental study, like most drug studies) are designed differently than drug studies. You also seem to confuse data with a study’s methodology. (I have been objecting to a lack of disclosure of the methodology and you have countered by talking about the data which is a different thing entirely.)
“I am basing my JUDGEMENT on the fact that it is old, and presumably time tested.”
But unless it was published in a reputable journal (and there is no evidence that it ever was, which I assume the authors would provide if it had been), no serious social scientist would bother to try to refute this study. Why? Because none of them would take this study seriously. There are literally thousands of unpublished (and by “unpublished”, I mean not published in a reputable journal) studies on any very narrowly-defined topic every year. There are at least dozens of good studies on any particular topic published in reputable journals every year. Why waste your time on all the dubious, unpublished studies like this one when you don’t have the time to even read, let alone support or refute, the many well-designed, well-conducted study are published in a reputable journals and which other social scientists will, therefore, take seriously?
Why might this study not have been published? I can think of three reasons. First, it might be poorly designed or improperly conducted. Those flaws would need to be so large as to render the results pretty much completely invalid and rule publication out.
Second, the study’s authors might not have the “appropriate” credentials (which IMO is not a good reason for a journal to not publish) and (more importantly) the authors could not find an appropriate academic (or other person with the appropriate credentials) to “front” for them. That is a troubling sign, especially where, as here, the study’s authors have already secured funding and some of that funding is from a reputable university. Usually that means that the study is either poorly designed or badly conducted. Otherwise academics would have a lot of incentive to latch on to the research. It’s already funded, the lackeys to actually conduct the research and write the rough drafts are already in place, and publication will help the academic in her/his career. So if it’s a reasonably good study, at least some reputable academics will jump at the chance to attach their name to a study.
The third reason is that the authors may not have been able to reject the null hypothesis, i.e., there is insufficient proof to believe that there were any differences between the groups. In this case, it might be that there was only evidence to be able to say that the wealthy whites were, in fact, any different than the other three groups. (It seems to me that such a finding might be considered too trivial to publish.)
Further, unless it were published in the 1980s (and AFAICS you have ZERO reason to believe that it was) as well as now, how can you say that it’s time tested? No one would have even known about it until now. I have no idea what you’re talking about when you say that.
“I do not have access to academic journals from the 1980s, so I can’t refute your point there. I would suggest that the name of the original study, and methodology, are under a different name. Again, write the authors of the paper.”
Why? There is ZERO indication that this study is valid. I have better things to do with my time than chase down dubious studies. It is the authors’ responsibility to show that the study is valid, not the readers to show that it’s not.
“I was a very small part (data collection and aggregation) of a long term cohort study and know the lengths that are taken when setting up one of these studies. They are built to last, this one has.”
You may know the lengths that were taken in the study in which you took part. But not all studies are conducted that way. You seem to be assuming without any evidence whatsoever that this study was designed in the same way and conducted in the same manner as the one in which you took part. I think those are very questionable assumptions.
The methodology of choosing the sample set was set in stone after the first study was done.
They then go out and survey the sample again every 5 years. They do not come up with another sample each time, the assume that the the sample from the original study was random, and representative. This is what you are saying isn’t true. I say it is. It is up to you to prove that it isn’t at this point and they only way to do that is to go back to the beginning.
The methodology on choosing the sample would have only been “new” to the original study. It would have been described in a very detailed fashion over the course of at least 50 pages.
This is just devolving into the global warming debate, everyone thinks that they should be provided with all of the data, and all of the conclusions along the way, but they don’t have the time to actually look at it.
Its there for you, you just have to find it. The fact that the data and methodology probably take up several thousands of pages is then used as proof that something must be wrong, without anyone actually bothering to look at each individual data point. Just like ‘climate gate’ then someone, who is not familiar with either the methodology of the study or statistics actually does looks into a small part of the study and finds ‘something wrong’, and uses that very small part to discredit the entire study. You don’t even have that very small part, and you refuse to even look for it.
Do your drug studies all give the sample set methodology and all of the statistical conclusions? No, they give you a one page summary of their findings.
Scientific and statistical studies of this size are a huge undertaking. They have lots of people working on them, over a very long time period.
In my mind this increases the validity of the study. In your mind, it decreases it. More data, over a longer period of time, is better.
When the authors published the latest study of the original sample set, they left out an explanation of what they consider to be settled ground(since 1984), the choosing of the sample set.
“The methodology on choosing the sample would have only been “new” to the original study. It would have been described in a very detailed fashion over the course of at least 50 pages.”
I agree with the first but question the second. IF the study were properly done, the sample would have been described in detail. So would many other things about the methodology. IF the study were properly done, the method would at least be summarized here and a cite given to the more complete description. That wasn’t done here. And there is no reason to assume it was done in the 1980s since if it had been, that study should be cited here now. Further, the authors would at least need to update the study’s methodology to show the new number of drop-outs, deaths, non-responders, etc. That wasn’t done here either.
You are again confusing the data (which I’ve never asked for but you keep talking about as if I have) with a description of the methodology (including how the sample was selected, how the data was collected, how the sample size was determined, how non-responders or drop-outs were handled, whether the traits were distributed normally or whether the distribution is skewed, bimodal, etc., what statistical tests were chosen and why (which would need to be done each time the study was published since the authors would need to explain why the statistics that they are using now are appropriate given the current sample size, skew, updated statistical methods, etc.). Any reasonably good study will describe its methods and give a citation to any earlier published data about the study. This study does neither. And your allusion to climate studies is a red herring since the ones people cite are those that have been published in reputable journals. This study has not been published in a reputable journal.
“Its there for you, you just have to find it” How do you know that? If it were there, the authors would cite it. I’ve never heard of an author not citing previous work done on an ongoing study. That would be deceptive and improper.
“Do your drug studies all give the sample set methodology and all of the statistical conclusions? No, they give you a one page summary of their findings.”
The reputable ones give you a reasonably good summary of the methodology and a place to write for more info. This study does neither. I’ve never seen a reputable study – drug or otherwise – that merely gives a one-page description of its findings. The synopsis may be one page (or even shorter) but the study itself is much longer.
“When the authors published the latest study of the original sample set, they left out an explanation of what they consider to be settled ground(since 1984), the choosing of the sample set.” But no reasonable study would do that. It would give a cite or link as well as a brief description of the methodology (and discuss any updates, etc. to the methodology). You also seem to assume that there was an earlier published study. I have never seen a follow-up study that did not cite all earlier studies. That would be improper. No reputable scientist would do that.
Further, there is no attempt to show if one should accept or reject the null hypothesis (i.e., the differences between the groups are probably real and not the result of statistical noise). Why do a study if you aren’t going to discuss that? There is no way of knowing if these differences are solely because of expected variances due to the statistical methods employed or if they represent actual differences between the groups.
“In my mind this increases the validity of the study. In your mind, it decreases it. More data, over a longer period of time, is better.”
Validity is a technical term that has nothing to do with my belief (or desires) or yours. If a study lacks validity, it’s because it has not met those requirements. I have nothing against longitudinal studies. They can be good or bad. If they are well-designed and well-conducted, I think they are often more helpful than non-longitudinal studies on the same subject. Contrary to what you claim, it does not decrease the validity of a study for me (or make me think less of it in any way or question it in any way merely because it’s longitudinal). However, not following established conventions in a field (without any explanation of the deviation or even noting the deviation) or especially not disclosing one’s methodology do lead me to question the study. Not bothering to publish in a reputable journal raises suspicion for me, especially where, as here the study was partially funded by the political offshoot of a major university. (See my reasoning, above.)
Apparently, failing to follow established rules for social science research or to show in any way that the null hypothesis was rejected doesn’t bother you. I guess we should just agree to disagree about that.
Dig here:
http://www.census.gov/compendia/statab/cats/income_expenditures_poverty_wealth.html
The term “African American” is incorrect and appalling.
Anyone who has ever taken the oath to become a naturalized citizen knows that you are to NO LONGER calls yourself by anything other than American. No hyphen either.
Blacks have to try harder if they want to increase the levels on this chart. Try doesn’t mean stand around and “Hope” something falls in your lap from Uncle Barack.
Try means working very hard to increase your education and making a better effort to assimilate. The prison culture has spilled into the ghetto so much that it is indistinguishable.
Ebonics is a way for blacks to keep from assimilating into the American culture, it is NOT a substitute for English.
Yves, really…
Could the trend PERHAPS be due to the fact that the study follows “the same cohort of families” – and therefore PERHAPS be due to the saving and investing habits of the families being studied? For instance, could it be relevant that African-Americans are less likely to invest in the stock market?
Ergo, one may assume, MORE likely to invest in other assets, for instance their home, which is not included? Or consumables, like cars, which lose value?
In any event, I wonder how you can attribute a trend that follows race, and not income, to “policies that favor capital over labor”, which “will, not surprisingly, widen the wealth gap.”
And I wonder, genuinely, how you can describe welfare culture and the war on drugs – the major factors at play in many black neighborhoods – as “policies that favor capital.” Unless you define capital as “promiscuous young bachelors and bachelorettes”, in which case, policies that discourage marriage and two-parent families are, indeed, “policies that favor capital.”
I hear your axe grinding, but frankly I see little sense or reason.
I read the report, and it is pretty sparse on details and long on conjecture (Yves’s comment is also pure conjecture in this case).
Sounds like they followed 2000 families over the study period. Whether they started with some larger number they don’t say.
They classify the families according to their incomes at the start of the study period. Because they did not keep track of variations in income over the course of the study period, we can’t see if the black families’ failure to build net worth was due simply to having less income, as would be the case if the primary wage earner suffered long periods of unemployment while the white earners stayed steadily employed.
They also don’t seem to have analyzed the families’ budgets to see if the black families simply spent more and invested less, despite (perhaps) having similar incomes throughout the study period as their white peers.
So basically this is a garbage study like most others on racial economics.
Nobody seems to have the courage to make a real effort to understand WHY problems like the one “studied” here exist.
Karen,
It is pretty evident from your comments that you are a bigot who is very wedded to the thesis that blacks overspend and who also knows nothing about the how to construct and interpret longitudinal studies.
The fact that the report is not detailed does not mean that the study is not well designed. Faulty reasoning.
I guess you didn’t fill in a census form this year. It very clearly asked what race you are and included African American or somesuch. It also included HIspanic and some other designations.
I would hazard to guess this is mostly a result of the lack of black family formation while most white families have 2 very close in pay workers. 2 bread winners = buying a better house which leaders to greater wealth growth thanks to the housing bubble.
Good work! Your post is an excellent example of why I keep coming back to read your excellent quality content that is forever updated. Thank you!