Bureaucratic Reengineering

A troubling trend in developed nations is that over time the share of the GDP devoted to government spending tends to increase over time. Here is a graph of U.S. Federal Government spending over time, (too bad it ends in 1997 – it would be going way up past 2000).

One reason this might occur is that the technology for government-produced goods is not keeping pace with the technology for private sector goods, and as we wealthier, we spend an ever-increasing portion of our budget on relatively scare public goods. Note the public goods are scarce only because they lack a good technology for producing them efficiently.

So over time, the government sector becomes relatively inefficient and drags down growth in the economy. The way to prevent this growth decline is to improve the efficiency of government-produced goods so that they too grow when the rest of the economy grows.

But reforming bureaucracies is easier said than done. Bureaucracies resist change. The type of employee that tends to migrate to the bureaucratic sector of the economy is the type that really doesn’t want to innovate and rock the boat. These people tend to be comfortable with the status quo. If someone comes in with bright new ideas, the bureaucracy might produce some valid objections, some invalid ones, and a lot of indifference, but little in the way of support.

It isn’t practical to suggest simply eliminating the bureaucracy. Often, the bureaucracy is doing a few important tasks along with some less important tasks. Some organization will need to do those tasks in any case.

Another approach to reforming bureaucracy is to thoroughly understand what essential tasks the bureaucracy performs and find a better way to do those tasks. Outside contractors could help design a new bureaucracy that performs all of the essential tasks that the old bureaucracy did with fewer employees and at a lower cost. Then the government could create this second bureaucracy in parallel to the original bureaucracy. In this way, the transition from one bureaucracy to the other will be smooth.

I can give an analogy with a road. It is very hard to repair a road and let traffic flow at the same time. Perhaps it would be easier just to build a parallel road and then, once the new road is finished, destroy the old road.

I am not claiming that this would eliminate government inefficiency and make the public sector as efficient as the private sector. I am only suggesting that we could improve the efficiency of the government sector enough to prevent it from being a serious drain on the growth rate of the economy.

The first bureaucracy I would reform is the CIA.