The efficiency effectiveness trade-off
Warning: technical, has personal anecdote.
I had to pay my hostel fees today. The ATM is halfway between my hostel and my college. I could have just gone to college, as I normally do, and withdrawn the money on the way home. Additional distance traveled: 0 meters. However, I might forget, or get lazy, or end up being in college later than the office is open – and thus end up not paying my fees. So I walked to the ATM, withdrew the money and paid my fees. Additional distance traveled: 400 meters.
The first method, picking the money up on the way home, was more efficient. The second was more effective.
Let X be some task. It is indivisible i.e. it is either done or not done, and nothing it between*.
A method M is said to be more efficient than N if it requires fewer resources to perform task X.
A method M is said to be more effective than N if the probability that task X will be performed is higher under M than under N.
Missing the efficiency-effectiveness trade-off is one of the intellectual vices of libertarians. The Japanese ministry of industry – MITI – is one of those government organisations that are credited with being effective, if not efficient. Markets may under-supply certain goods – education, for example – relative to the optimum, due to credit constraints. They are not effective in supplying the optimum quantity of the good. Governments may then try to supply it, but end up wasting resources relative to the private sector production of the good. To be fair, many libertarian policies – school vouchers, for example – try to make the market effective by using efficient methods (cash transfers instead of directly supplying goods and services).
*It is not necessary that X be a task or that it be indivisible. For example it could be the optimum quantity of a good supplied.