Blockchain and Pricing in the Mortgage Insurance Industry — The Great War to Détente
By Michael A. Verlezza
The mortgage insurance sector is an interesting animal. In existence since the 1880s, competition on price was heretofore impossible — a handful of firms used a rudimentary signaling game canonical example of a signaling game to ensure oligopoly pricing. This model changed, however, as firms recently began to roll out so-called “opaque” pricing models. Rather than price insurance policies solely on creditworthiness, equity, and the lender’s desired coverage level, more variables were introduced. Rate cards are no longer published, and new prices now blend additional dimensions, such as the actuarially justified lender type (credit union versus national bank, for example) with more normative factors (such as proprietary regional economic forecasts). It is useful to think of these pricing models as extremely lengthy linear models, with the coefficients therein serving as the “secret sauce” by which an insurer could again compete on price.
In the interest of maintaining the old order of price parity, firms now spend considerable resources identifying the rates imposed by competitors. As Federal regulations bar insurance companies from soliciting quotes from one another directly, field sales staff ask lenders to tell them what prices exist in the marketplace, while business analysts comb public filing databases for whatever insight may be available. This manual process is labor and time intensive, and is not merely inefficient from a business standpoint, but perpetuates uncertainty around price. Given the demand for mortgage insurance and the well-documented troubles around government entities Fannie and Freddie, opportunities for improvement must be identified and implemented.
For the moment, we set aside the obvious benefits associated with smart contracts, and instead look solely at the potential for blockchain to create a dynamic optimized market clearing price.
At the philosophical core of blockchain is the egalitarian notion of fairness — that the distributed ledger creates transparency and equal footing not generally available through other means.
While the aforementioned robust (some may say “oppressive”) regulatory framework has indeed provided a measure of protection against discriminatory pricing practices, the more robust opaque pricing models mentioned earlier are engineered, in large part, to circumvent these customer protections.
Consider then a distributed ledger of a firm’s mortgage insurance contracts. Embedded in each transaction is the rate assigned, the purchase price (and thus the premium collected). Furthermore, baked in are additional loan attributes, such as Metropolitan Statistical Area, FICO score, and household income. The current stock of insured houses serve as the genesis block, and with every payment made and with every claim paid, the ledger is updated to reflect the current state of play. Coupled with a rudimentary machine learning model, risk models may be tested and prices modified to reflect actual loss and expense ratios.
Further, envision a scenario where the genesis block is not limited to data from a single firm, but rather the book of business from all five mortgage insurance companies. Now, we are able to capture the entire market, marrying the regulatory framework designed to prevent against usurious pricing and ensure sufficient reserves with which to claims.
Is not difficult to see where our thought experiment is heading. With visibility into the entire marketplace, our (well-trained with existing data) machine learning mechanism guarantees that each new policy written receives the optimized market clearing rate. This rate exists at all times and may be applied to every loan, as risk is captured at a perfect or near-perfect rate, and thus price is engineered at binding to guarantee both profitability and solvency (vis-à-vis the capacity to pay claims).
Not only does this approach highlight the scalability and efficiency so often touted in the blockchain community, but it clearly presents the more egalitarian approach from the perspective of both insurer and insured.
Note however, that one person’s industry-wide optimized pricing engine is another person’s cartel. Furthermore, these are insurance companies operating in an industry with 130 years of history. Thus, regulatory approval and adoption would be the long pole, rather than implementation. These are technical problems with known solutions, but it is natural to expect each firm’s reluctance to buy in, as such a process is anathema to the current — albeit obsolete — paradigm.
Having said that, it is not unreasonable to anticipate a future with two different types of firms: those which embrace blockchain, and those which go dark. It is my sincerest hope that forward-leaning executives will see the value in eliminating overhead, the certain reputational windfalls, and the transparency blockchain so readily affords them.
ABOUT THE AUTHOR
Michael A. Verlezza is a Fulbright Scholar, having conducted research in game theory and comparative economic policy at Queen’s University in Kingston, Ontario. His research has touched an array of fields, including finance, healthcare, human resources, and academia itself. His interest in Blockchain and Cryptocurrency goes back to 2012. An associate of the AEIOU, Michael resides in Raleigh, NC with his wife and two daughters.
The AEIOU is the independent think-tank of The Adrealm Foundation, creators of a blockchain-based digital advertising ecosystem. The observations and commentary published on AEIOU-curated platforms and channels do not necessarily reflect the point of view of The Adrealm Foundation. We welcome thoughtful, well-written articles that are directly or conceptually related to blockchain technolgy, cryptocurrencies, and ad-tech.
For more info about the submission, you can contact us here: laura@adrealm.com.
