Friday, March 5, 2010

Time Horizons and Political Preferences

On this blog, I've repeatedly explored the idea that political preferences are a result of an individual's time horizon. My contention is that liberals are more focused on the short-term and conservatives the long-term, especially when it comes to economic ideas.

I often refer to Kenyes famous quote:
In the long run we are all dead.

But he also writes this in The General Theory of Employment, Interest and Money:
Moreover, life is not long enough; — human nature desires quick results, there is a peculiar zest in making money quickly, and remoter gains are discounted by the average man at a very high rate.

I wonder how much The Big Assumption comes into play. Might it be that Keynes is projecting his personal discount rate onto others and drawing incorrect conclusions? Tough to tell. Finding the median discount rate is an empirical question, although I'm not quite sure how one would go about it.

Anyway, many political economy models assume things about the interest of labor that I think are shifted toward the short-term in a very subtle way. Often they assume that workers want lifetime employment and high wages. They then shift to labor unions as the political expression of those preferences. Prima facie, this seems to hold, but my skepticism of unions has always made me uncomfortable with this transition.

What workers really want is to maximize the net present value of their earning over their lifetime. If I were to model it, then it would be the sum of wages discounted according to their personal discount rate. However, economically there is an inflation versus unemployment trade-off that I think sums up most people's political preferences on economic matters.

So, you have to not only discount the sum the wages by each worker's individual discount rate, but also the rate of inflation. Plus, you'd have to factor in a value for unemployment. This would result in something like this:

The sum of: (1 - Median unemployment rate) * [ (Predicted Wage) / (1 + personal discount rate + median inflation rate) ^ n ]

Where n is the number of years from when a worker starts working.

The personal discount rate is determined entirely by the person. However, the median unemployment rate and inflation rates are a product of political choices. They can't be predicted precisely. They are normal distributions around some mean, but, and this is the important part, that mean is determined politically. Therefore, a worker has partial control or at least political preferences for those to values. Unemployment and inflation are inversely related. A lower median inflation rate means, broadly speaking, a higher unemployment rate.

Getting to my point, my thesis is this: the personal discount rate is the independent variable in this equation (from a political perspective) and preferred median unemployment and inflation are dependent variables.

What's the relationship between discount rate and inflation/unemployment? The lower the discount rate, the more people are more willing to accept a higher probability of short-term unemployment if it means higher growth in their personal wages over the long-term and a lower inflation rate to eat into those future wages. And this translates into political preferences for more laissez-faire policies.

Part of the problem with unions is that they are unable to change employers or sectors. What I mean by this is that a union always represents the workers of an specific employer or a specific sector. The union, as an organization, is stuck whereas its members aren't necessarily. In the industrial economy, where there was lifetime, or at least long-term employment, this discrepancy didn't matter. But when there is no lifetime employment or people see changing jobs and even sectors as a way to maximize their lifetime earnings, then the unions preferences, as an institution, for more job security and higher wages (i.e. lower unemployment/higher inflation) act at cross purposes with their members.

I'm just trying to hash all this out and see if I can find some sort of model to understand it with.

BTW, an important assumption for this to hold is that countries that are more willing to accept cyclical unemployment (like the US) grow faster and that this translate into higher wages over the long-term.

UPDATE: The more I think about this the more I'm convinced the key point is how the inflation versus unemployment trade-off affects long-term economic growth, and more particularly the growth in wages. I'll have to look into that. If anybody sees anything else, then let me know.

Update II: So, I realized that I just put predicted wage in, but no way to predict it over time. Moreover, I think the key empirical question would be whether a higher variance in unemployment is positively correlated with wage growth.

Also, this omits the idea of worker as capital owner. Given that many retirement plans are invested in equities, workers have interests to protect their savings after their working life. And rather than just wages we'd have to include their income from capital in their political preferences.

I still think discount rate plays the primary role. I'm just not quite sure of the mechanism.

UPDATE III: I can't believe I forgot about taxes. Logically speaking, a person with a high discount rate would value more current spending now far higher than they would worry about increased taxes at some unspecified point in the future. They are already discounting their future wages so much that it increasing taxes on them wouldn't be that big a deal.

No comments:

Creative Commons License
This work is licensed under a Creative Commons Attribution-No Derivative Works 3.0 United States License.