Home > data science, modeling, musing, open source tools > Quantifying the pull of poverty traps

Quantifying the pull of poverty traps

January 16, 2013

In yesterday’s New York Times Science section, there was an article called “Life in the Red” (hat tip Becky Jaffe) about people’s behavior when they are in debt, summed up by this:

The usual explanations for reckless borrowing focus on people’s character, or social norms that promote free spending and instant gratification. But recent research has shown that scarcity by itself is enough to cause this kind of financial self-sabotage.

“When we put people in situations of scarcity in experiments, they get into poverty traps,” said Eldar Shafir, a professor of psychology and public affairs at Princeton. “They borrow at high interest rates that hurt them, in ways they knew to avoid when there was less scarcity.”

The psychological burden of debt not only saps intellectual resources, it also reinforces the reckless behavior, and quickly, Dr. Shafir and other experts said. Millions of Americans have been keeping the lights on through hard times with borrowed money, running a kind of shell game to keep bill collectors away.

So what we’ve got here is a feedback loop of poverty, which certainly jives with my observations of friends and acquaintances I’ve seen who are in debt.

I’m guessing the experiments described in the article are not as bad as real life, however.

I say that because I’ve been talking on this blog as well as in my recent math talks about a separate feedback loop involving models, namely the feedback loop whereby people who are judged poor by the model are offered increasingly bad terms on their loans. I call it the death spiral of modeling.

If you think about how these two effects work together – the array of offers gets worse as your vulnerability to bad deals increases – then you start to understand what half of our country is actually living through on a day-to-day basis.

As an aside, I have an enormous amount of empathy for people experiencing this poverty trap. I don’t think it’s a moral issue to be in debt: nobody wants to be poor, and nobody plans it that way.

This opinion article (hat tip Laura Strausfeld), also in yesterday’s New York Times, makes the important point that listening to a bunch of rich, judgmental people like David Bach, Dave Ramsey, and Suze Orman telling us it’s our fault we haven’t finished saving for retirement isn’t actually useful, and suggest we individually choose a money issue to take charge and sort out.

So my empathetic nerd take on poverty traps is this: how can we quantitatively measure this phenomenon, or more precisely these phenomena, since we’ve identified at least two feedback loops?

One reason it’s hard is that it’d be hard to perform natural tests where some people are submitted to the toxic environment but other people aren’t – it’s the “people who aren’t” category that’s the hard part, of course.

For the vulnerability to bad terms, the article describes the level of harassment that people receive from bill collectors as a factor in how they react, which doesn’t surprise anyone who’s ever dealt with a bill collector. Are there certain people who don’t get harassed for whatever reason, and do they fall prey to bad deals at a different rate? Are there local laws in some places prohibiting certain harassment? Can we go to another country where the bill collectors are reined in and see how people in debt behave there?

Also, in terms of availability of loans, it might be relatively easy to start out with people who live in states with payday loans versus people who don’t, and see how much faster the poverty spiral overtakes people with worse options. Of course, as crappy loans get more and more available online, this proximity study will become moot.

It’s also going to be tricky to tease out the two effects from each other. One is a question of supply and the other is a question of demand, and as we know those two are related.

I’m not answering these questions today, it’s a long-term project that I need your help on, so please comment below with ideas. Maybe if we have a few good ideas and if we find some data we can plan a data hackathon.

  1. January 16, 2013 at 7:26 am

    This also applies to countries (e.g. Greece)?


    • January 16, 2013 at 7:46 am

      Presumably, but that would be even hard to quantify considering the myriad factors involved.


  2. TJ
    January 16, 2013 at 8:32 am

    Great post, great project. The judgmental quality discussed in the Times article (also criticized by Felix Salmon) is a important piece of the trap and, unfortunately, long a part of western culture. Sadly, I watched the poverty trap happen to a musician friend over the last couple of years. In addition to struggling financially, he began to believe all of the negative judgmental stuff and became convinced he was a failure in all aspects of his life. He was harassed constantly by a debt collector over relatively small amounts, struggled with his family over his career choices, lacked real health care coverage and became convinced it was all his fault. The judgment enhances the toxic environment, forces bad choices, and destroys a person’s confidence. Very painful to see – he was a sweet, extremely talented guy.
    Whatever laws are in effect, or could be in effect, are not enough on their own to counteract the toxic judgmental environment that goes along with the poverty spiral. People in this situation need more access to information and groups that want to provide support without exploiting them further.


  3. KH
    January 16, 2013 at 9:01 am

    If you really want to perform this experiment, you could buy a bunch of people’s debt and, instead of forgiving it like the Rolling Jubilee, redo the terms instead of forgiving it. Then once you have the debt, don’t act like a debt collector. Of course, the practical and ethical issues with such a scheme abound, and it’s not obvious that you would be able to separate effects. (Are there other debts which are being collected via harassment? How do you control for extent of indebtedness? Family situation? Previous experience with debt collectors?)

    Perhaps a more viable alternative is studying the effect of the Rolling Jubilee on people. Specifically, the feedback loop on the models is probably not going to be broken by the forgiveness of debts. (Especially since RJ reports forgiveness to no one, the models can’t get the data.) So you can at least compare people on the bad end of the model loop with people who are on the bad end of the model *and* debt loop. Not exactly the experiment you were going for, but it would be relatively doable and ethical and would still answer a question.


  4. Zathras
    January 16, 2013 at 9:05 am

    Great topic here. These issues are very poorly understood. I have one suggestion here. As with any data mining problem, it’s best to engage a subject matter expert as soon as possible. The question here is who that would be. Academics? Maybe, but there likely won’t be many out there. 20 years ago state consumer protection agencies might have been good sources of information, but in many places these agencies are so emaciated now that they likely can’t provide much information here. Other possible sources of information are attorneys. There are attorneys who specialize in predatory lending cases, either on an individual or a class action basis. These attorneys are likely the only people you could ever talk to who have seen internal documents of lenders and give you the real view on the ground.


    • January 16, 2013 at 9:10 am

      Great idea, but why would they give us their data?


      • Zathras
        January 16, 2013 at 9:26 am

        Because your analysis could give them back better insight on the nature of cases, what to look for, damages exploration, etc. They could benefit from the qualitative insights as well.


  5. Justin
    January 16, 2013 at 11:09 am

    Great topic. This paper, “The real Costs of Credit Access: Evidence from the Payday Lending Market” by Brian Melzer is I think currently the best research I’ve seen looking at the comparison of outcomes in states with and without payday-loan bans: http://qje.oxfordjournals.org/content/126/1/517.short (sorry not sure where to get an un-gated copy). His conclusions “loan access leads to increased difficulty paying mortgage, rent and utilities bills”


  6. Heather
    January 16, 2013 at 11:36 am

    I think there’s evidence that internet availability/useability is far lower for people below a certain tax bracket. That may affect whether or not your thoughts become moot in the near future. Many sites are not easily accessible from mobile devices, and many people in the “poverty cycle” use only their phones for access. Payday loan companies will probably figure that out and work around it, but just thinking out loud.


    • January 16, 2013 at 11:37 am

      Yes that’s a good point. The research needs to be firmly anchored in what people actually do.


  7. NotRelevant
    January 16, 2013 at 2:05 pm

    “people who are judged poor by the model are offered increasingly bad terms”

    That could be just regular finance at work… the idea is that to make a normal profit, interest received back must cover all costs, including the cost of principal that isn’t returned. However, there are numerous examples where these markets are inefficient, like when collateral loans include very high interest. These are often called hard money lenders or predatory lenders.

    Also included are pawn brokers, but I’ve seen enough crap for sale in the pawn shops to know they are the first people to see not just stolen property but broken, worthless stuff too. I have a cousin who was a pawn broker for a while… he had a river of new toys (guns, tools, steroes, fishing reels, cars), new jewelry, new girlfriends, and plenty of cash too. But even he got taken a time or two.

    I have years of education and am heavily invested in degrees, but would I prefer to have taken all that money and invested it in a pawn shop? Not really. All that education just helps me realize that a pawn shop can be a very good business. And they lend to people without being judgmental.


  8. ZHD
    January 17, 2013 at 12:56 am

    I like your take on this. If only you had the patience and narcissism to be a behavioral economist, the field might move along a bit more swiftly.

    Here is a podcast with the author of the study (discussing the study) you’re talking about, you’ll get a better idea of the methodology and intentions from this: http://whyy.org/cms/radiotimes/2011/12/21/the-psychology-of-poverty/

    Limiting myself to using existing cognitive biases:

    I think the “poverty trap” scenario is unremarkable. In controlled scenarios of heightened scarcity (i.e. a scenario in which there are not enough available resources for all players), each actor begins with an internal model that assigns a greater-than-normal value to the commodity that needs to be acquired. These actors also have a fitness function in which having less than adequate resources is bad, and having adequate resources is good.

    Research into the pseudocertainty effect (from who else but Kahneman & Tversky) shows that people are willing to take riskier bets to avoid higher negative outcomes. Once an actor puts on leverage to obtain a commodity, that triggers the confirmation bias of the other actors that “yes indeed, this is highly valued and required leverage.” Once leverage becomes the norm, a Gresham’s dynamic has been created and actors make moves based in reactance bias.


  9. alagator2k13
    January 17, 2013 at 12:45 pm

    Cathy, speaking of rouge models, I’m wondering what you think of this:


  10. msl
    January 18, 2013 at 11:54 am

    I think the broader point about models leading to higher interest rates [or any onerous condition] leading to the poor being worse off which conforms to the models expectation is actually fairly well-covered in economics. Check out Stiglitz’s papers on information asymmetry and interest rates 1979-1981, then swap the word “information” for “model” and you’re already mostly there. The feedback loop aspect [where the model makes it’s own prediction more true] is called performativity [see the book, Engine, not a Camera] and has been written about extensively with respect to financial markets but has yet to be written in a sufficiently mathematically coherent form that would be published in QJE. It also has not been applied to humans [rather than financial assets] yet – but perhaps it ought to be!

    Commenter Ramsey Haddad points out that the same effect may apply to countries too. He is correct, check out Krugman 1989-1991 who writes extensively on the “laffer curve” for countries that higher interest rates rationally assigned can lead to lower revenue for debt-collectors because the probability of default rises so quickly.

    Of course, there is a massive econometric literature on this too.

    Of course, the behavioral dressing applied to so-called poverty trap effects is also interesting [in that it exacerbates the already existing effects of a rational, information deficient market].


  11. Nathanael
    January 20, 2013 at 6:20 pm

    There’s a reason why high interest rates used to be banned outright. There’s a reason *all* debts were cleared in bankruptcy (yes, including student loans).

    Without the option of “clearing the decks” with bankruptcy, at a certain point, it becomes rational to keep borrowing and not even attempt to accumulate wealth, because there’s no real opportunity to. And usurious loans are the “fast track” into the debt trap.


  1. No trackbacks yet.
Comments are closed.
%d bloggers like this: