Open source lending tinged with racism

Open source lending tinged with racism

Summary: If open source lending is to work out the problem of racism must be fixed.

SHARE:
TOPICS: Open Source
11

Prof. Utpal Dohlakia of Rice UniversityWhether described as peer-to-peer lending or open source lending, there is no doubt that personal lending auction sites like Prosper and Kiva are doing big business.

Think of it as a reverse eBay. Instead of describing a product, you describe your financial self and how you will pay the money back. In return users with money bid on your loan, and sign a standard contract.

For the first time we have a real academic research paper on how this is all working out. Utpal "Paul" Dholakia (Bombay '93) from Rice's Jones School of Management has studied Prosper and the results aren't surprising.

Racism abounds.

The borrower characteristics having a negative effect on auction funding success are, as anticipated, debt-to-income ratio and being a minority (African American).

When banks operate this way we can sue them. When individuals, on their own, make decisions tinged by racism, you can't force someone to make a loan.

But here's an idea. Competition. If sites are organized around specific types of lenders, or the search for specific types of deals, these problems may start to disappear.

In other words, minority investors pool resources just as Prosper does and go to market seeking good opportunities among minority borrowers.

Dr. Dohlakia wanted to see if open source lending democratizes the process. To an extent it does. But to another extent it doesn't.

So here's another experiment the Doc might want to try. Make two applications for the same loan, one under the name Paul Dolak, using the picture above, the other under the name Utpal Dohlakia using the picture on the Jones School site.

If open source lending is to work out the problem of racism must be fixed. How would you fix them?

Topic: Open Source

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

11 comments
Log in or register to join the discussion
  • Profit opportunity

    If some non-financial factor is restricting the ability of some people to obtain the loans, then anyone who ignores that factor will be able to charge a higher rate than would otherwise be obtained.

    If someone's financial profile would justify a 5% rate (just to name a number) and simple reluctance prevented any takers at that rate, then there's an opportunity to obtain a premium by loaning at 6%.

    Software could identify these disparities between appropriate and available rates, with the likelihood of greater profits than would be obtained by normal rules.

    Of course, once one person has followed that procedure others will be attracted by the results. With enough people looking for a premium, the premium will diminish until it disappears.

    This market correction doesn't apply with red-lining because banks have an appropriate reluctance to lend outside areas in which they have experience. No competition waiting to use an opportunity. But the internet can remove geographic factors.

    Key is the ability to evaluate factors sufficiently well to determine an appropriate rate with precision. Because that's not entirely a science, there's still a risk involved.
    Anton Philidor
    • Good talkback. Thanks

      The problem is that the actions which seem rational and normal at Prosper would be illegal if applied by a regular bank. This includes redlining, from which I suffered for years.
      DanaBlankenhorn
  • Inaccurate Reporting

    The report "THE DEMOCRATIZATION OF PERSONAL CONSUMER LOANS?" did not find that racism is an important negative factor in the success of peer-to-peer loans and certainly did not indicate that "racism abounds." In fact, the report seems to indicate the opposite.

    The authors clearly state that demographic factors have little if any effect on the likelihood of funding success. They write, "The strongest direct effects belong to the effort measures (especially the provision of detailed explanation regarding borrowers' needs and their ability to repay the loan), and the financial indicators (especially, the better credit grades.) Demographic characteristics, such as gender, race, and marital status have little effect (if at all) on the likelihood of funding success." And, "Specifically, better credit grades strongly affect borrowers' stated starting interest and requested amount, and the provision of detailed or general explanation affect borrowers' starting interest rate. Borrowers' demographic attributes [i]play a little or no role at all in affecting the three decision variables[/i]. The African American race is one important exception, which will be discussed later." (emphasis mine)

    In that later discussion, the authors are unclear as to why African Americans are the only exception but they do NOT conclude it is due to racism. They suggest it may have financial motivations. "As discussed earlier, the regression of starting interest rate on borrower characteristics did show that African Americans start their auctions with a significantly lower interest rate than other borrowers. This may make African American borrowers' auctions less attractive to potential bidders when compared to auctions listed by others."?

    The authors do suggest that these factors aren't strong enough to account for the discrepancy, they don't conclude racism is at play (particularly given that no other race has this issue). While they state "The borrower characteristics having a negative effect on auction funding success are, as anticipated, debt-to-income ratio and being a minority (African American)", the causal factor seems to be financial viability as the largest determining factor in the closure of a personal loan. The authors continue, "The provision of a picture, which affects negatively on auction funding success, is not independent of credit rating--borrowers with high-risk credit are more likely to provide a picture, hence the effect is non-surprising."
    The point is that it is credit rating, not race per se that is the determinate. While there is a correlation between race and credit rating, the paper does not indicate that lenders are not rejecting borrowers specifically because they're African American. The data indicate that finances are the important factor.

    More importantly, the abstract of the paper states clearly that demographic factors play a much smaller role in P2P lending than it does in institutional lending which seems to contradict the claim and intent of this article. The authors write, "Although demographic attributes such as race and gender affect likelihood of the auction's funding success, [i]their effects are small when compared to those of borrowers' financial strength and effort when listing and publicizing the auction[/i]. These results are substantially different from the documented discriminatory practices of regulated financial institutions in the US, suggesting that individual lenders bid more rationally when their own investment money is at stake." (emphasis mine)

    One final point. There are four authors of the paper. Utpal Dholakia is one of them.
    ppardi@...
    • As usual with topics like this...

      ...the truth lies at the opposite end of the article than the headline, in more ways than one.
      JohnMcGrew@...
  • This has been happening off-line for decades.

    "If sites are organized around specific types of lenders, or the search for specific types of deals, these problems may start to disappear. In other words, minority investors pool resources just as Prosper does and go to market seeking good opportunities among minority borrowers."

    Many minority communities have informal lending pools, especially among immigrant communities. No, I'm not referring to loan sharking. There are informal systems where those who have succeeded lend to new residents.
    CharlieSpencer
  • RE: Open source lending tinged with racism

    Please don't malign Open Source by calling this "open source." "Peer to peer" is a slightly more apt borrowed term for it, but again it's not really right. Why isn't there a better term than "exchange community" for sites (whether Kiva or Bookmooch) that enable people who need something to match up with people who have something?
    Walt_z
  • Please don't malign Open Source

    Please don't malign Open Source by calling this "open source." "Peer to peer" is a slightly more apt borrowed term for it, but again it's not really right. Why isn't there a better term than "exchange community" for sites (whether Kiva or Bookmooch) that enable people who need something to match up with people who have something?
    Walt_z
    • Yeah

      What the heck this type of lending has to do with open source code is beyond my understanding. ZDNet should either explain or apologize.
      schmandel@...
      • I noticed..

        that this Dana journalist has the potential that could easily beriddle Mr. Spock for eons.
        TedKraan
  • Unregulated lending tinged with peril

    Notwithstanding my perspective of having been in banking and lending for the last 30 years, the events in the financial services industry since last autumn's fall of Countrywide Finance have led me to the conclusion that more regulation is needed in specific areas, for example derivatives trading.

    I see many positive aspects to these collaborative lending sites and I would hope that they come into the fold of best practices instead of being a foundation for later troubles like subprime mortgages.
    schmandel@...
  • RE: Open source lending tinged with racism

    How about investigating the most pernicious form of discrimination we have: uglism (discrimination against the ugly)? I would be willing to bet that an attractive black woman does far better on Prosper than an ugly, fat white guy. So why does it always come back to the tired old mantra of racism? I admit I discriminate. When I see a attractive, young woman, I usually stop and check out the listing. I try to hold myself to my objective lending criteria, but I know that in the back of my mind, a pretty face influences me a bit.
    ZDNET_guest666