Problems with the Policy Ineffectiveness Proposition

So what's the Problem?

Remember, we associated series A, the unbiased series, with Rational Expectations, and Rational Expectations, in turn, is a basic assumption of the Policy Ineffectiveness Proposition. But the evidence tells us that changes in production are more like series B -- they are highly permanent. The most permanent kind of series is called a "random walk." The random walk is a series in which this year's observation is equal to last year's observation except for a small difference that is absolutely unpredictable -- that is, "random." Aggregate production approximates a random walk pretty closely.

Using the language of the Reasonable Dialog again, this observation would undercut the Policy Ineffectiveness Proposition pretty badly, except for one thing. As we have already pointed out, the evidence is that people's expectations are unbiased. And we said the highly permanent series was biased. The evidence seems to be on both sides of the issue! But, of course, that can't be right. Let's look one more time at Figure 2:

Figure 2: Long And Short Run Aggregate Supply with Highly Permanent Production

The problem is not really that the production dots are highly permanent. The problem is that they are mostly to the right of the LAS. That means that people are finding them surprising. But if production is very permanent, and people have Rational Expectations so they know that production is very permanent, then people would not be surprised by the relatively large production.

Perhaps the problem is with what we have called the "hidden assumption:" the assumption that the NAIRGDP doesn't change as a result of changes in Aggregate Demand and Short Run Aggregate Supply.

There is another possibility. Let's look at another example of highly permanent changes in production.

Changes in Production