In recent months I have had to deal with numerous activities that involved the use of my FICO score: buying a new car, shopping for a good deal on auto insurance, applying for student loans for my kids, getting a new credit card. A few years ago I was the victim of identity theft, as a result of which I signed up for a service that alerts me to any changes in my credit score. It has been fascinating to see how my behavior impacts my FICO score: applying for new credit? Score goes down a few points. Deciding to pass on a new credit card and instead start using an old card I hadn’t used in a few years? Score goes up a few points (someone will have to explain this one). Sometimes just racking up a large balance during the month because of business travel will trigger a transient change, even though I pay the balance in full every month. Fortunately for me, these small changes are not creating problems, though if I were, for instance, about to get a mortgage, I might find that even a swing of a few points could put me into a different risk bracket, which could translate into thousands of dollars because interest rates depend so heavily on your FICO score. Clearly, the folks at Fair Isaac Corporation put a lot of thought and complexity into their algorithms. Equally clearly, the notion of a single algorithm that attaches a single score to an individual is highly questionable. For instance, I was looking for a new credit card because I was carrying both personal and business expenses on the same card, which is great for the points, but makes expense reports a nightmare and was costing me time and money. Since my balance was fluctuating a lot each month, and FICO penalizes high balances on a single account more than lower balances on multiple accounts, you would think that adding a second credit card would have made me, if anything, a lower risk. However, applying for that second credit card made me appear more risky. Estimating the true credit risk of an individual is a difficult problem. Unfortunately today’s conservative approach is hurting a lot of people who are working hard to stay afloat or even rise above their current state. The system is rigged so that the less advantaged you are, the more disadvantaged you become: a person who can afford to borrow money easily might save a few hundred dollars each month by getting a better rate, the same hundreds of dollars that might prevent a hard-working lower class individual from being able to afford that home. The interconnections between credit score, risk, cost of capital and social success is a highly complex problem that is artificially masked by the apparent simplicity of a single numeric score like FICO. The irony of it all is that the current conservative attitude of financial institutions is the result of the hubris and risk-taking behavior of the very same financial institutions in the last decade. This conservative attitude is having a substantial impact on our economy and is contributing to the growing divide between the “haves” and “have nots” in our society. A more dynamic approach to credit risk estimation that is centered on individual behaviors and causality should yield even better risk estimates and avoid some of these unfortunate problems. Photo Credits here and here But the real problem is that financial institutions of all types, still reeling from recent travails, seem to have become overly reliant on numbers for their decisions. A couple of years ago I applied for a modest mortgage and, to my surprise, was denied by two large lenders. When I spoke to a banker at a local Credit Union, he explained that alimony and child support count as consumer debt, which made my “bottom debt ratio” look horrible (bottom debt ratio is equal to the total housing expense plus debt payments divided by the gross monthly income). Ironically, when I asked how my ratios would look if he took my salary net of child support, and then calculated ratios based on my “net” salary and my other expenses, he said my ratio would have been great and I would have no problems qualifying. So instead of building equity for myself, I am paying rent and helping my landlord build his equity.
The Complexity of Credit and the Futility of FICO
On the Adoption of New Drugs: A Two-Part Series
Part 2 – One Reality
In a previous post, I briefly described the drug adoption contagion myth: the landmark Tetracycline study does not offer evidence that drug adoption spreads through social networks. But it is a widely-held belief and pharmaceutical companies still allocate a significant amount of their marketing resources based on this premise. Our experience is that the contagion effect, if it exists, tends to be small. Furthermore, opinion leaders (determined either through unreliable self-reporting or somewhat more reliable sociometric measures) have little impact on the speed of adoption –in other words, there are no real influencers. To be fair, however, our experience also suggests that each drug follows its own adoption path, which depends on a lot of factors such as the disease itself, the market at the time of introduction, and how it is prescribed and used.
On the Adoption of New Drugs: A Two-Part Series
Part 1 – The Myth
In 1954, pharmaceutical giant Pfizer was interested in determining how physicians decide to adopt a new drug so that it could more effectively market its products through detailing and traditional media. By knowing how physicians acquire reliable information and who they trust, Pfizer could market its new drugs more effectively, optimizing the allocation of marketing resources among detailing, media advertisement, continuing medical education, etc. They funded a landmark social network study aimed at showing the effect of interpersonal influences on behavior change in relation to the adoption of Tetracycline, a powerful and useful antibiotic just introduced in the mid-1950s. Pfizer hoped tetracycline would diffuse rapidly because it was a tremendous improvement over existing antibiotics. The Pfizer-funded study contained two major advances over previous studies in that it relied on a behavioral measure of time of adoption by looking at prescription records and used network analysis to identify opinion leaders. However, numerous subsequent studies of this work revealed a number of weaknesses in the collection and analysis of the data and the study is inconclusive: the uptake in tetracycline adoption cannot be assigned with confidence to social network effects.