Wednesday, October 29, 2008


Greg Mankiw's Blog: My Personal Work Incentives

Greg Mankiw's Blog: My Personal Work Incentives

Another thought on incentives. Imagine a hypothetical world where everyone's tax rates were 0% except for one unlucky person who gets hit by the 100% rate. Using Mankiw's analysis, this world would be a much more horrible place than it is now or would be under either the Obama or McCain plans.

But it would be just the opposite. 99.9999% of the population would enjoy the massive $28 incentive for earning an extra $1. The one unlucky guy could even be well compensated if just a portion of the population were willing to pitch him a buck for drawing the short stick in the tax lottery.

Which leads us to a flaw in his analysis here. He focuses only on the worse case scenario rather than examining the entire population. For example, for most people the estate tax is 0% because their estates will not be more than the exemption. When confronting the issue of whether to push a bit harder to make that extra $1 they don't care about Warren Buffet's incentives, they care about their own.

To really measure the incentive impact of the different tax plans one has to divide the population into groups and calculate an incentive rate for each one. Computationally challenging but not impossible and it provides a better picture than simply looking at the worse case.

UPDATE: Greg Mankiw has covered this before here. However the calculation does not seem to take population into account. In the above hypothetical the average tax rate is 50% (100% + 0% = 100% /2 = 50%) but that's a very different economy than one where half the population is paying 0% and the other half is paying 100%.

Tuesday, October 28, 2008


Ceteris Paribus Is Now Always Your Friend

Over on Pseudo-Polymath I was having a discussion over whether or not Obama is a socialist (or, ‘more socialist’ than McCain). I made the point that McCain’s health care plan is estimated to cost $2T while Obama’s is $1T.

McCain’s health plan is $0 you know. What sort of plans of his is he going to be able to push though a Democratic locked Congress?

Here we meet the old friend of economists, ceteris paribus or “all else being equal”. There are real relationships between variables in our world. If you raise the price of something, people will buy less of it and vice versa. But lots of stuff is always happening. You may raise the price of something and end up with more sales because people have suddenly decided it is popular, or because something else is sold out and this is the next best thing or any one of an uncountable number of possible scenarios.

So we usually say, without actually saying it, that if everything else stays the same then something will happen if a certain other thing happens. Raise the price and you’ll get fewer sales. If the price of milk skyrockets, coffee demand will probably fall (most of us don’t drink it black) and you can probably come up with a million other examples.

In this debate the response is a bit unfair. Of course neither candidate’s plan will ever be enacted as currently written. If McCain, through some odd chance, wins the election he will face a very angry and very Democratic Congress and even if he is serious about pushing his healthcare plan, it almost certainly would end up dead on arrival or drastically changed in its final form. But if we are going to ask who is ‘more socialist’ we should try to hold everything else constant. What is Obama’s plan, what is McCain’s plan and let’s try to measure ‘how socialist’ they are. If we are going to try to predict what type of policies will actually be passed 1, 2, or 3 years from now we are going to have to play a lot of speculative games. McCain, for example, might agree to a ‘very socialist’ health plan in exchange for Congress not yanking the funding for an extended Iraq occupation. A President Obama might have to ditch healthcare if the financial collapse makes gov’t borrowing impossible. (Don’t get me wrong, I think this whole socialism debate is nonsense. In less than a month we just decided to spend almost a trillion dollars to buy up banks, bad debt and bail out huge companies….the ‘socialism’ argument at this point is a bit like the arguments over Star Trek versus Star Wars...except the people who get impassioned about that are typically more reality grounded).

But Ceteris Paribus isn’t always our friend. Over on Greg Mankiw’s blog, he has a fascinating post comparing work incentives under Obama versus McCain’s tax plans.

Basically he looks at a no-tax world. Assume you work today to feed yourself and your family and provide a standard of living that you find acceptable. You would take on extra work (an hour of overtime, a side job, etc.) primarily to benefit your kids by being able to leave it to them. If you earn $1 extra today, he estimates that it would benefit his kids by $28. Why? No taxes plus compound interest.

Under McCain’s plan the benefit is $4.81 and Obama’s it is $1.85. This is, of course, due to taxes. Some of this, though, is a stretch. For example, the estate tax would only be relevant if you’re estate was going to be larger than the relatively generous exempt amount. This assumes you do not shelter your earnings either as unrealized capital gains or inside a 401K/IRA but the general point is valid. You do suffer a lower incentive with Obama’s plan than with McCain’s and that is a real cost.

But what is even more fascinating is the ‘no tax’ case where the incentive is a whopping $28 (nearly 6 times as much as with McCain’s supposedly ‘market friendly’ plan). Why not simply propose no taxes?

Here is where Ceteris Paribus isn’t so much of a helpful friend. We during the ‘good years’ of the Bush administration we ran up over a trillion dollars of debt. In the blink of an eye, at the last moments of the Bush years, we just added another $1-3 trillion (depending on how you count the bailout combined with various ‘guarantees’ by the Federal Reserve and Treasury Dept).

There are two sides to the incentive to earn that extra $1. The first, of course, is the relatively easily measured tax implications of earning that extra $1. The arithmetic might get a bit complicated but it’s basically simple stuff. The second, though, is exactly how easy is it to earn that extra $1? Is spending 3 weeks looking for work as good as hopping on Craig’s list and discovering the local store is willing to pay you to spend a night helping them do their year end inventory? Certainly not and that has nothing to do with tax rates.

The question then is assuming we adopt a policy of lowering tax rates, what will that do to the economy over the long run after we have been aggressively adding to our debt and spending? If you answer is ‘everything will be great’ you are either totally deluded or you have discovered the ‘free lunch’ (hike government spending, drop all tax rates to zero and everyone will be happy). The non-existence of the free lunch is one of the few truths economics has that is non-debatable.

Mankiw, though, creates an artificial free lunch by the cleaver deployment of ceteris paribus. Because ‘all else is held equal’ the economy says nothing as tax rates are lowered as the government absorbs trillions in debt. All we are left with is asking ourselves do we prefer to have more incentive or less incentive…if only life were that simple.

This page is powered by Blogger. Isn't yours?