Resilient Communities are the Foundations of a Resilient America.

Guest Post: Credit Score Model for Measuring Community Resilience by Brian Dabson

Continuing with the recent theme of measuring resilience, Brian Dabson of the University of Missouri’s Institute of Public Policy asks  – can the model we use for credit scoring be applied to community resilience.  A provocative question, and an interesting answer. – mjp

A credit score[1] is a numerical summary of a customer’s apparent creditworthiness and is intended to provide a lender with an indication of the likelihood that the borrower will default on the loan.  The higher the credit score the more likely the loan will be paid back in full and on time.  The lower the credit score the greater the risk of default.  Credit scores are widely used to make decisions on whether to lend and on what terms for mortgages, auto loans, and credit cards.  A high score extends the range of products available and at the better rates; a low score narrows the range of credit available and significantly increases interest rates.

Scores are based upon information gathered by consumer reporting agencies, the three main nationwide ones being TransUnion, Equifax, and Experian. These agencies gather, organize, and disseminate consumer credit information: the amount of a loan, the balance on a credit card or loan, the payment status of the account, and any judgments and bankruptcies.  The agencies prepare a report on every consumer, which for some can be quite extensive and detailed.  To enable the process of credit review to be efficient and consistent, the agencies apply an algorithm or scoring model to convert the information in the reports into a single summary score.

There are several different scoring models used, but the most widely used is the FICO Score. This is calculated from several categories of predictors based on information in the credit reports.  There are five broad categories: payment history (35 percent), amounts owed (30 percent), length of credit history (15 percent), new credit (10 percent), and types of credit (10 percent). The percentages reflect the weighting or relative importance of each category.

Credit scores have enormous importance for lenders. They are used:

  • In the pre-screening process to determine who will receive what kinds of credit offers
  • In the mortgage application process to determine eligibility, terms, interest rates, points, and fees
  • For assessing suitability for auto loans, the maximum amount of loan, the interest rates, and if a co-signer is required
  • For credit cards to determine eligibility, interest rate, credit limit, fees, and rewards.
  • In account reviews to raise or lower limits on credit cards, or in offering new or different products.

Not surprisingly, credit scores are also of vital importance to consumers as they determine the availability and cost of credit.  The good news for those who have low scores is that they are able to improve their scores over time by changing their financial management behavior.  Specific actions in relation to the five predictor categories will “repair” bad credit, raise scores, and expand credit options at better terms. Such actions might include paying bills on time or keeping credit card balances low.

What if the same ideas could be applied to measuring community resilience? Can we imagine a universal system where investors could access a scoring system that would provide them with an indication of the likelihood that their investments would be better protected in the event of a disaster or significant threat? And what if communities could see clear short-term benefits in terms of access to investment capital, lower interest rates and insurance premiums by taking actions to make themselves more resilient in the longer-term?

Credit scoring models are based on massive amounts of detailed data drawn from millions of individual credit files, categorized in ways that have been found to be good predictors of creditworthiness.  They rank order consumers by predicted credit risk so that the score shows whether a consumer is more or less likely to repay a debt relative to other consumers. A resilience scoring model would need to draw upon consistent community data organized in ways that predict the capacity and readiness to withstand levels of shock and subsequently to return to the same or higher levels of functioning. Each community would be assessed according to a set of economic, social, environmental, and infrastructure indicators, and assigned a score (or scores) that reflect their comparative resilience against other communities in their state, across the nation, or those with similar levels of inherent vulnerability.

Communities would have access to data, tools, and playbooks that enable them to improve their resilience score over time.  The incentives to seek improvements would, as suggested above, be financial – with expanded individual, business, and government access to credit and insurance at reduced rates being the goal – but also psychological, with greater civic pride and ability to attract jobs and people to a safer, more caring community.

The challenge in devising a resilience scoring model is to identify indicators that actually predict increased resilience so that they can be used with a high degree of confidence by investors, while at the same time represent factors that communities can do something about to improve their resilience.  A coastal location or susceptibility to certain weather conditions may be important factors in assessing risk, but the best a community can do is to recognize the risk and take action to mitigate the impacts.

The advantage to investors, whether bond financiers, business lenders, banks, or insurance companies, would be that they could better assess and price risk, and over time, their experience would provide further data on the predictive ability of the indicators, so that they these can be fully integrated into the investment and credit system.

Differential pricing based on resilience would inevitably penalize development in flood prone areas, exposed coastal zones, and forestlands, as well as giving strong signals to communities without adequate building codes, planning and zoning regulations, disaster preparedness plans, and adequate infrastructure that their inaction comes at a cost.  This may prove unpopular in some quarters but it will be necessary to get the nation on the track to increased resiliency.


[1] This description is drawn from the Consumer Financial Protection Bureau’s Report to Congress, The impact of differences between consumer- and creditor-purchased credit scores, dated July 19, 2011, and from the FICO Banking Analytics Blog, www.fico.com/analytics, accessed July 27, 2014.

An afterword…Brian’s piece speaks to the “dirty little secret” of community resilience – communities generally aren’t interested in resilience unless there is an incentive.  My perception is that insurers are starting to get the message and are trying to figure out what kinds of incentives make sense.  However, most lenders still have not recognized that a natural disaster can wipe out their investments just as effectively as poor business practices.  There are occasional glimmers of movement in this direction, indicating some momentum, but clearly it is only halting progress, at best.  Thanks, Brian!