## Saturday, February 11, 2006

### Supply-Side 101

One of my six regular readers emailed me with a request: to educate him on supply-side economics. This post is my attempt at that…

At the very heart of supply-side economics is an assumption about the way that individuals and companies behave. In almost any system or theory of economics, you’ll find the notion that demand is influenced by price. Basically this says that you’ll sell more of any good or service if you lower the price — the lower the price, the more consumers will make the decision to buy. I don’t know anyone who quarrels with that assertion, most likely because it is everybody’s personal experience.

There’s a related notion that not everyone understands as well: that it isn’t necessarily better for the seller to sell more of something. Perhaps a simple example will make this obvious. Suppose you went into business selling lemonade on the street in front of your house. We’re keeping this very simple, so further suppose that the only costs you have are lemons, sugar, and plastic cups — and that those costs total 10 cents per glass of lemonade. Not knowing what price you should charge, you run an experiment: you try four different prices (15 cents, 20 cents, 25 cents, and 30 cents) for one week each, and record the resulting sales (250 cups, 170 cups, 140 cups, and 90 cups). Which week is best for you? If you do the math, you’ll find that your profit for each week was \$12.50, \$17, \$21, and \$18 — so the best price for you (the seller) is 25 cents. You’ll sell lots more at 15 cents, but you make so little per cup that you’re worse off. So as the seller, where do you set the price? If you went into business in order to make money, then of course you’d set it at 25 cents.

If you made it through that, you might be wondering what the heck this has to do with supply-side economics! Well, it turns out that this is the heart of the matter. Most likely, as you read the example above, you accepted the notion that the number of glasses of lemonade you sold varied according to how much your customers had to pay. Of course, right?

This is exactly what supply-side economics says about the behavior of people and companies as tax rates change. If you think about it, from the consumer’s perspective tax rates are just a different price for whatever good or service they’re buying. Let’s take one example to make it concrete: long term capital gains tax. This is a tax on the profit you make by selling any security (e.g., stocks) you’ve owned for over a year. Buy a stock at \$10, sell it two years later for \$12, and you’ve got \$2 in capital gains — which the federal government taxes you on. When you are deciding whether to sell a stock you own, do you take that tax into consideration? Well, you’d have to be pretty darned dense not to! I know that every time I’ve ever sold stock, I paid very close attention to that (blankety-blank) capital gains tax! So doesn’t it stand to reason that if the capital gains tax is lowered (effectively lowering the cost for you to sell stock) that more people will sell stock? Also the converse: if you raise the capital gains tax, don’t you think that fewer people will sell stock? Not only is the preceding common sense, it’s also borne out by experience: every time the capital gains tax has been reduced, stock turnover went up, and every time it has been increased, stock turnover went down.

If you absorbed the preceding, then (in technical terms) you have dismissed the “static model” of tax economics. The static model (which our government still uses for many forecasts!) assumes that nobody will change their behavior because tax rates change. That model seems to me to be just plain stupid. Click on the chart above right to see a graph of this. The red line shows the results of a static model: as tax rates increase, so does the tax revenue. Even as the tax rate goes to 50%, the revenue keeps going up — as if people would really behave this way! The static model’s main advantage is that it gives liberals a reason to screech about tax reductions. This only works if we’re all stupid sheeple. Oh, don’t get me started on this!

A simple alternative to the static model, and one that often comes first to mind when thinking about this, is the “linear model”. In this model, the assumption is that consumer behavior does change as price (or tax rates) change — and it changes in proportion to the change in price or tax rate. In other words, if you drop the price by 10%, then 10% more people will buy. If this was reality in the tax world, then it wouldn’t matter what tax rate you set — the revenues to the government would be the same. How boring is that? The problem with the linear model, however, is that pesky reality. People don’t really behave like this. Let’s take an example to illustrate: suppose that the price of a movie ticket was \$10, and at that price 100 people watched the movie at your theater. Now cut the price in half, to \$5 — what do you suppose would happen? Most likely, you’d get more than double the number of customers, because \$5 seems like a major bargain (by today’s standards, anyway!). Now try the proportional opposite: double the price to \$20. Do you think half as many customers will show up? I rather doubt it — \$20 is so much higher than the competition that practically everybody will go to them. This is very non-linear behavior, and it’s quite the norm in the real world.

There are many ways to model non-linear behavior, and economists and mathemeticians debate them endlessly. But for the purposes of this discussion, these debates are all details that don’t matter, since all the models have a more complex, “curvy” something like the green line on the chart. The non-linear models (and reality) all have something in common: there’s always a “sweet spot” where the most profit (or tax revenues) will be made; a certain price (or tax rate) where they are maximized. In the example I’ve used, tax revenues are maximized at around a 6% tax rate, considerably lower than the 15% starting point. The sweet spot could just as easily have been on the other side of the current rate, at a higher rate — it all depends on how those tax rates affect people’s behavior.

The premise of supply-side economics, translated into tax policy, is that the government should never set tax rates higher than that sweet spot. At the sweet spot, not only are tax revenues maximized, but it’s a healthy, vibrant economy that’s causing it. Setting tax rates higher than the sweet spot means lower economic activity — not good from anybody’s perspective — and less tax revenue. That’s not good even from a liberal’s perspective, though getting one to admit that is pretty darned hard! Setting tax rates lower than the sweet spot isn’t necessarily bad — while it reduces tax revenues, it encourages economic growth in whatever area of activity is being taxed. This sounds like something a liberal would reflexively reject, and a libertarian reflexively embrace <smile>. For example, suppose we’re talking about the tax on the purchase of solar cells for powering homes, offices, or factories. Suppose the sweet spot on that (for maximizing tax revenues) was 5%. To encourage the adoption of solar power, one might consider reducing that tax, even to zero — despite the reduction’s impact on tax revenues.

Another key element of supply-side economics is what I’ll call the “regenerative effect”. This is another common-sense notion that virtually every economics system or theory embraces: the idea that economic activity has an impact greater than the immediate participants in the activity. For example, if I grow some corn and sell it to you for \$10, then you and I have each benefited. But I’m going to take my \$10 and do something with it (I’m going to buy a big box of chocolate!). The guy who sold me the chocolate now has \$10 to do something with. The same thing is true for companies: the profit they make regeneratively increases overall economic activity, to everybody’s benefits. Liberal interpretation of corporate profits often diverges from this; listening to them, and reading their writings, I get the idea that they believe corporate profits all end up in the hands of evil, filthy rich “capitalists” (voiced in disdainful tones) who, presumably, bury it all in the cellar — because in their view it’s certainly not doing the world any good. This is a remarkably ignorant mindset, so obviously wrong-headed that I presume it to be an article of faith, rather than a reasoned position.

You might also wonder how (for example) outsourcing might affect supply-side economics. Well, really outsourcing is a completely separate issue — except so far as tax rates affect the rate of outsourcing. If you assume (and I most certainly and emphatically do not!) that outsourcing is bad, then as a matter of policy one could lower tax rates in the affected industry to make our domestic suppliers more attractive. But this really has nothing to do with supply-side economics; one could do the same thing even under traditional liberal economic theory…

That’s the lesson for today!