Drawing an ethical line in the big data sand

However it’s defined, the concept of big data ultimately boils down to money. Consumer companies, for example, want to improve marketing efforts by targeting specific products at their likely buyers. All companies want to maximize internal efficiency to save on power bills and production processes. Even non-profit and government organizations want to eliminate costly defrauding of their systems and figure out the best ways to spend their limited resources. But is there a line we shouldn’t cross when it comes to using analytics to squeeze every last dollar out of an operation?

I think there is, and I think it’s at the point of transition between things we want and things we need. Case in point: On Tuesday, the New York Times published an article describing how specialized software is letting landlords maximize rents by analyzing supply, demand and market prices to suggest the highest-possible price to charge for a particular apartment. It sounds logical enough if you’re in the property business, only peoples’ homes aren’t plane tickets, hotel rooms or rental cars.

Real people, real consequences

Rent-pricing is only one problematic application of analytics, but it’s illustrative of an ethical dilemma that might also apply to things such as medical services or groceries. Demand for rental units is skyrocketing because of the foreclosure crisis and because banks aren’t lending even to some who could afford homes at today’s low prices. In an empathetic society, the answer would be to make it easy for displaced families and young individuals just setting out on their own to secure apartments and maybe some sense of stability. In a capitalistic society, apparently, the answer is to seize on the opportunity to gouge them for every nickel and dime.

One housing-industry executive told the New York Times reporter, “Now it’s more about the analytics. Maybe this person is paying 15 percent below market, and as much as we love them, maybe it’s time to let them move on and capture that 15 percent.”

Heading into the holiday season, this strikes me as a perfectly Scrooge-like viewpoint. Only for most families, there won’t be a happy ending in their cases, because Scrooge is hiding behind algorithms that insulate him — perhaps as part of corporate policy — from caring. The only logical outcomes are that when leases expire and renewal rates skyrocket ARM-style, renters who can’t afford them move into worse or smaller apartments, or leave the area altogeter. But people aren’t chattel: studies have shown that frequent moving has can be traumatic for children and can result in lower quality of life later on.

Automatically setting rent prices at the statistical ceiling also could have an broader economic impact. Renters who can afford the new rental rates, but just barely, could end up stressing our welfare system when they didn’t have to before. That extra $200 a month could have been the grocery budget or the heating bill, or it could have been gas money for the daily commute. Now they’re looking at food stamps, taking a lower-paying job closer to home, or both. Maybe they’re not going to the doctor early enough because they can’t cover the co-payment, and everyone pays when they end up with a huge emergency room bill.

How much sheer profit is justifiable?

But algorithms don’t take these externalities into consideration. They only act how they’ve been programmed to act — analyze every unit against the relevant factors and tell a landlord how much it can get away with charging. I don’t think anyone’s under the impression that property management is an altruistic endeavor, or that renting property should be a profit-free business, but how much profit is enough? This isn’t like corn prices rising in response to drought or higher production costs — higher rent just because someone will pay it is pure profit.

And claims, such as that in the Times article, that this type of price-optimized system can actually benefit renters ring hollow to me. Accepting less-favorable lease terms in order to secure lower rent is hardly a deal, and it’s just not realistic that most renters will have the market knowledge or math skills to successfully to haggle on things such as start dates and varying rent prices (something that didn’t work out so great with mortgages). The management company, after all, has a computer telling it the minimum it can accept and still meet revenue goals.

I bang the big data drum as loudly as anybody, but I think there’s a line — however fuzzy — between what’s possible and what’s ethical. Letting the numbers decide where and when to place online ads, how much a hotel room should cost, or even to assess credit is one thing. It’s yet another when the numbers alone start directly displacing people from their homes, or otherwise deciding who can or cannot access the staples of survival.

Feature image courtesy of Flickr user Kheel Center, Cornell University.