I see more of these every week: we should build X because it will create a zillion jobs or we should build Y because it will cause economic growth or we should do Z because it will save a ton of money. Does anyone ever go back and look at past predictions to see how accurate they were? If they did, they’d see that not only are such predictions usually wrong, they are horribly horribly wrong. Predicting the economic consequences of government spending makes astrology look like physics. If you want to study a classic case, look at the projections for Medicare spending back when Medicare was enacted.
So, why is this? I think three factors contribute to this mess:
- The world is complex and our models are simple. We simply haven’t developed tools that are powerful enough to make prediction reliable.
- Technological advances change everything. The original predictions for Medicare couldn’t account for heart transplants, angioplasties, or a thousand other things that have dramatically changed medicine and life expectancy.
- Human behavior changes. One of the advantages of getting older is witnessing how much ordinary behavior changes over time. Some of this is in response to technology (think cell phones) and some is a gradual cultural shift (think race relations).
So, what to do? It’s not like we can just decide not to plan or forecast. But we can decide to be very skeptical and only commit to things when we think the case is overwhelming. It may be
better to only bet on the sure thing.