Perverse incentives and inequality

An interesting post, Is This Why Americans Have Lost The Drive To "Earn" More by Tyler Durden. Durden is looking at the aggregate net impact of a multitude of well-intended and important public policies (health, income, housing, etc.) and identifies that there is a range between $30,000 and $65,000 of income where it is technically worthwhile in the short term to not work any harder; you are as well off or better off relying on various government support programs than if you were to work harder.

See the horizontal dotted line, marked as the Highest Welfare Cliff, in his graph.

I don't know whether his numbers are rigorously right but the challenge of perverse and unintended incentives is by no means an unknown problem, though it has garnered more attention in the past in Europe (with its more comprehensive and robust social programs) than in the US. The fundamental issue of mixed and poorly understood incentives is familiar to the private sector as well. Years ago I had a telecom client and their problem was that they were experiencing major ebbs and flows in monthly demand for different types of services (and the consequent impact on revenues). Some of their services had much better profit margins than others, so this lack of predictability was an important issue.

The root cause was traced to the sales incentives for different services that were being pushed by the customer service/sales area. Each division of the company had their own programs for promoting different services (voice, internet, cable, cell, etc.), all programs uncoordinated with one another. The sales guys knew to a penny their net benefit from the different incentive programs and adjusted their behaviors accordingly. Each division was constantly tweaking their sales incentives with the consequence being that the sales guys were constantly changing their focus based on the net benefits to them. Hence the unpredictability - no one was looking at the whole complex picture.

So it is not surprising therefore that there is a problem of negative incentives to personal productivity. The question is what can be done about it. Just as you can't be half pregnant, you can't be partly motivated over a life cycle of productivity. The more you coast at the beginning, the less you are likely to achieve at the end. Cultural values will carry many people across this disincentive gulf. They will educate themselves and their children, work hard, and save for the future because that is the right thing to do for them, regardless of the incentives of government programs. But some unknown portion of the population will behave like the incentive-based sales force. They will seek to optimize their current "income" based on the net impact of various government programs, regardless of the long term impact on one's personal productivity.

This is yet another illustration of the fact that when it comes to the Human System (complex, chaotic, homeostatic, hidden feedback loops, etc.), you can't rely on good intentions but instead have to focus on actual outcomes. In this instance, not only are you undermining the types of behaviors that are the foundation for long term prosperity (both for the individual and the community) but you are also exacerbating other problems unintentionally.

Durden is probably correct with his numbers. What needs to be examined is the extent to which those counter-productive incentives have an impact on people's behaviors. I suspect that it is material though hopefully not catastrophic. If the counter-productive incentives do indeed have a real and material impact on people's behaviors, you likely have at least two undesirable consequences.

First is the more obvious one - if people respond to the tactical incentives and reduce their personal effort, they are likely reducing their long term strategic productivity; to their and the the country's detriment.

The second consequence is perhaps less obvious. If there is this broad mire of disincentive towards hard work between the income ranges of $30-65,000, then you likely will drive people to either end of the distribution. Some people will abandon themselves to low income choices because of the support they receive from government programs which shield them from the full consequences of those choices. They become captive clients of the state with little prospect of escape. In other contexts, people would be outraged at this malevolent disenfranchisement.

Other people, recognizing the mire but committed to their own prosperity, will work that much harder to enter the higher levels of income where one can once again benefit from one's own efforts.

As people, responding in their own fashion to the perverse incentives, migrate to one end of the spectrum or the other, you will necessarily have a thinning out of the middle. This movement to the extremes from the middle would mean that the gap between the wealthiest and poorest would widen.

The increase in the income gap between the bottom and top quintile in the past forty years is well established (though it is unclear whether the widening has any sort of real world consequence). Regardless of whether widening income gaps are important for any reasons other than a residual sense of "fairness", people have struggled to explain exactly what has caused the increasing gap. Globalization, outsourcing, offshoring of manufacturing, changing demographics, changing family structures, the transition to a knowledge economy, etc. have all been advanced as the mechanism that explains increasing income inequality.

What Durden's graphs and numbers point to is the possibility that increasing income inequality is a result of government policy.