Greg Mankiw has a nice op-ed on dynamic scoring
The issue: When the congressional budget office “scores” legislation, figuring out how much it will raise or lower tax revenue and spending, it has been using “static” scoring. For example, it assumes that a tax cut has no effect on GDP, even if the whole point of the tax cut is to raise GDP.
My thoughts.
1. I am against dynamic scoring. Dynamic scoring means using an economic model. I think that politicians and the press give too much credence to economic models as it is. Even static scoring requires some modeling, but the modeling has more to do with spreadsheet arithmetic as opposed to claiming to be able to predict economic behavior.
2. To the extent that the CBO has to predict economic behavior, I think it should present several scenarios, as opposed to a point estimate or a range. Cochrane says it well:
It’s a fact, we don’t know the elasticities, multipliers, and mechanisms that well. So stop pretending. Stop producing only a single number, accurate to three decimals. Instead, present a range of scenarios spanning the range of reasonable uncertainty about responses.
Responding to another point from Cochrane, Mankiw writes,
you need to specify how the government is going to satisfy its present-value budget constraint. You might be tempted to ask the model what happens if the government cuts taxes and never does anything else. But you won’t get very far. The model will tell you that the government has to do something else eventually, and it won’t tell you what will happen if the government tries to do something impossible.
What I hear Greg saying is that to properly do dynamic scoring, you would need to include a model of future policy responses. That is a point well taken, but I am not sure that I would restrict those policy responses to be only doing things that are possible. Policy makers are doing impossible (that is, unsustainable) things now. The challenge is to predict the outcome of undertaking unsustainable policies until you cannot do so any more.
Of course, the traditional “static” scoring does not solve the problem of how to predict the outcome of unsustainable policies, either.
I share skepticism of dynamic scoring and agree with Cochrane’s post, but yoru claim here seems to overstate things:
“Even static scoring requires some modeling, but the modeling has more to do with spreadsheet arithmetic as opposed to claiming to be able to predict economic behavior.”
‘Static’ scoring *is* a prediction of economic behavior. It’s just (probably) not the correct one. We should not confuse static projections for being in some way model-neutral or model-free. They themselves embody, and are impossible without assuming, one particular possible path of future economic behavior – growth, employment, tax receipts, etc. That’s a model like it or not. (It’s similar to an investor being all in cash thinking he’s not making any trades. Yes, he is.)
It’s a fair question to ask whether the static path is the most likely or informative one along which to make these calculations. I won’t pretend to have a good answer to that.
Some good questions to ask:
1. What would dynamic scoring of Greece has looked like, say, 15 years ago.
2. Were they doing ‘the impossible’?
3. Assuming the answer to question 2 is plausible ‘yes’, then what should the Greek CBO equivalent have assumed about future conditions and policy responses.
4. By what rational process could the Greek CBO have arrived at the most reasonable set of assumptions about those future policy responses?
Another question could be asked about the Disability Insurance Trust Fund, or the Post Office’s finances. The DI trust is on schedule to be completely exhausted sometime next year and, (in)conveniently, leading up the presidential election. By any reasonable use of the term, Congress’s requirements to pay a steady level of benefits combined with the instructions to make use of the trust and current revenue streams is ‘impossible’, and the impossibility of which is about to be proven in the very short term.
Again, what are we supposed to assume about what the political system is going to do about this, and does the CBO arrive at those conclusions regarding what happens when one of the trains of the system keeps chugging along towards the end of a fiscal cliff?
The whole system has a lot of trains heading towards a lot of cliffs, and that just compounds the problem by creating the necessity for a greater multiplicity of response assumptions.
The “magic” is in that word of **will** in “what will happen . . .”
It may be worthwhile (to some limited extent, given politics & lobbying) to consider and lay out in differing formats:
What can happen.
What might happen.
and the “whys” of each.
Doesn’t CBO scoring take place when the proposal is nearly ready for vote and thus a done deal? So, it basically serves political purposes. Therefore it is most likely that the wider ranges offered will be used politically. We have even seen versions of this when we were told “it is good to spend more now because we are in a recession” or “it is good that people are voluntarily incentivized out of the work force because it is voluntary.”
Well, you can be sure “dynamic scoring” will be done in a way to please whoever is in charge. Probably a lot like “static scoring” is done today.
CBO has used dynamic scoring for years–for spending changes. The rule change is about dynamic scoring of micro effects such as tax avoidance.