1. I think most people understand the supply and demand less than we think. That’s because supply and demand are more confusing and complex than what is taught in mainstream EC101 lectures today. Maybe you don’t think so, but I do. Suppose we asked students who have already taken EC101&102:

“Oil prices fell. For this reason, people will probably buy more gas(as if something is cheaper, people will likely buy more.) (True/False).”

What do you think will be the response of many students? Most students would say “true” because “The demand curve slopes downwards; so the quantity demanded increases as the price falls.” But is it so? When I took EC101, I had the opportunity to observe many students. Almost all of them pass the course at the end of the semester without knowing more than they knew before. That’s why student essays consist of an endless cycle like “prices fall, demand increases; prices increase, supply increases.”

To be honest, I think economics is taught the wrong way. Have you looked at the introductory chapters of the best-selling textbooks? In the textbook next to me, all I see are meaningless examples such as “After an unexpected hail squall, the supply of crops decreases and prices increase,” “Unexpected high interest in Dean Martin concert driving up prices.” Students knew this even before taking the EC101. Do you know someone who has never studied economics and has never heard of the phrase “supply and demand sets the price”? And what most of the students know exactly at the end of the semester is nothing more than that. Basic econ courses should teach students intuition and interpretation, instead, they learn nothing but memorizing things like the “5 factors that shift the demand curve.” “If price increases, supply increases; when demand decreases, price increases; when X happens, the supply curve shifts to the left/right.” And it goes on like this.

This is an example of “reasoning from a price change,” and it’s the first thing many students learn in the intro to economics. Heck, it’s not surprising if you consider that even most professors make this fallacy all the time.

Let’s go back to the first example and take a closer look. Oil prices plunged, and oil consumption fell too. So low oil prices may also reduce oil consumption. How? Draw a supply and demand curve. What happens if the demand curve shifts to the left? When the price falls(if we only know that), there is the same chance that the quantity will increase or decrease.

Scott Sumner makes the same point in one of his posts: 

I often ask the following question to upper level econ or MBA students who have already taken principles:

Question: A survey shows that on average 100 people go to the movies when the price is $6 and 300 people go when the price is $9. Does this violate the laws of supply and demand?

Very, very few can answer this question, especially if you ask for an explanation. Even worse, I think there is a perception that there is something ‘tricky’ about this question, something unfair. In fact, it is as easy a question as you could imagine. It’s basic S&D. It’s merely asking students what happens when the demand for movies shifts. I cannot imagine a less tricky question, or a more straightforward application of the laws of supply and demand. In the evening hours the demand for movies shifts right. Price rises. Quantity supplied responds. What’s so hard about that? And yet almost no student can get it right. Our students enter EC101 knowing one of the two things they need to know about S&D, and they leave knowing one of the two things they need to know about S&D. Maybe instead of having them memorize mind-numbing lists of “5 factors that shift supply,” and “5 factors that shift demand,” we should just tell them to read something that will explain what economics is all about, something that portrays economists as detectives trying to solve the identification problem, something like Freakonomics.

If there are any other economics instructors out there I’d like to know what you think. I really don’t think we need to teach students what happens when frost hits the Florida orange crop. Perhaps we should just put supply and demand into an appendix and tell them to study it if they need to. Instead devote 100% of chapter 4 to the identification problem. Leave all the technical stuff for students majoring in economics to take in their intermediate level courses. Or maybe the identification problem is too hard, and we should simply forget about teaching supply and demand. Devote the whole course to opportunity costs, incentives, marginal analysis, etc.”

Read the whole thing, then tell me supply and demand is not complex than what’s taught in college.

2. I don’t favor the IS-LM. It’s not because I don’t understand the technical mechanism of the IS-LM model. I will not argue that the IS-LM is “theoretically or technically” flawed. Instead, I don’t think that the reasoning economists have made from the IS-LM is correct. In other words, I think the IS-LM will cause economists to misinterpret certain situations. That’s why I almost always use AS-AD instead of the IS-LM(more precisely, I prefer using several models with a partial equilibrium perspective). First, I’ll start by saying that the IS-LM is a useful and well-working model for periods like the Great Moderation. But I believe this is due to a different reason than many people think(Taylor Rule).

I think the problem with IS-LM is that most people(even most economists) seem to think they can tell the stance of monetary policy and/or changes in monetary policy by looking at changes in short-term nominal interest rates. In contrast, short-term rates are only a variable that accompanies changes in the money supply relative to the demand for money. Therefore, changes in AD are driven by expectations for future changes in NGDP.

My problem with IS-LM is that while it may seem like a useful model for periods like the Great Moderation, it doesn’t work well during periods of high inflation, when inflation expectations change rapidly. Likewise, it does not seem to be a useful model in periods such as the Great Depression and the Great Recession, where expectations fall too fast. According to economists’ interpretations of the Great Depression using the IS-LM, monetary policy was supposed to be expansionary. Likewise, in 2008, many economists said that monetary policy was expansionary after looking at interest rates and the IS-LM. But it wasn’t. Hence the IS-LM can be useful when everything is stable, but otherwise it’s not a useful model. But why would we use IS-LM in our analysis if it is a model that is only useful in stable periods? I am not in favor of using a model that can mislead economists at critical times like 1929 and 2008.

One might argue that “the problem is not with IS-LM, but with economists misusing the model,” but I’m not making a technical objection to IS-LM anyway. Even if the IS-LM is technically a perfect model, it seems clear that economists were misinterpreting something in 2007. As I said above on supply and demand, low rates could result from a shift in the IS curve, but it’s clear that most economists didn’t think so in 2007 because they thought low rates implied expansionary monetary policy. That’s why I prefer a simple AS/AD model, where output can change depending on supply and demand shocks. And AD is equal to NGDP (determined by monetary policy), which is why I argued earlier that the concept should be called “nominal expenditure(s),” not AD. I think this model is more useful than the IS-LM model.

3. I don’t think corporate tax cuts benefit only the rich. On the contrary, it benefits everyone. Likewise, I think capital gains tax is meaningless. Wage taxes, in principle, tax current and future consumption at the same rate. On the other hand, a capital gains tax taxes future consumption at a higher rate than current consumption. What principle suggests that something like this would make sense? The new proposal will raise the rate above 50%. The highest rate in the world is 30% (Sweden), which is close to the current US top rate.

To be honest, I’d prefer to repeal income tax for both individuals and corporations(and inheritance tax) and replace it with a consumption tax. Then a VAT where the poor pay 0% would be ideal. Likewise, I’d prefer a progressive payroll tax. And a progressive property tax. And, of course, the carbon tax.

This tax system would be more efficient without requiring the high rates we have now. Perhaps one could argue that the consumption tax will be regressive, but that’s not true:

Consumption tax is almost all about being consistent over time. We can have a tax system that is more progressive but simple, useful, and efficient.

4. One of the reasons I prefer the term “nominal expenditure(s)” is because it refers to a specific amount. That’s why I think AD is actually something different from “demand.” For the same reason, I think using different models for business cycles will give more precise results. I prefer to use one model for nominal shocks and a separate model for everything else (real shocks). The second includes the natural rate of unemployment, technology shocks, taxes, etc. It will help to explain the changing factors. These are especially important when prices are flexible.

But the sticky wages/prices problem makes it impossible to explain business cycles by using only real shocks. Indeed this problem is very different from the rest of economics(which is why I think DSGE models are unreliable). The model I would prefer to use for nominal shocks would focus on explaining changes in nominal aggregates. That’s a model that focuses on employment fluctuations attributed to price and wage stickiness. The “optimal level” will vary depending on the type of wage stickiness, requiring us to define nominal shocks in terms of wage stickiness. This is why nominal expenditures are more useful than “AD”: when we talk about nominal shocks, we will ultimately focus on wage stickiness. Instead, many macro models focus on variables that do not affect the nominal optimal level.

Sticky prices, nominal shocks, and the optimal level are key variables in explaining business cycles. So when I saw Sumner’s model for a few days ago, I thought it was pretty close to the model I’ve been considering for a few years. It’s not exactly the same, but it almost is:

 When I used to teach macro, students had all kinds of trouble understanding AS/AD, partly because they assumed it was basically a supply and demand model.  Thus if I asked them to show the effect of population growth, they might shift the AD curve to the right.  In fact, population growth shifts the AS curve to the right, causing deflation (as we saw during 1870-95).

Why were students confused?  Probably because in microeconomics an increase in population really does shift demand to the right.  More consumers out shopping, etc., etc.  But aggregate demand isn’t really about demand at all; it’s about money.  More specifically, it’s about the medium of account.  In the late 1800s, gold was the medium of account, the thing in terms of which prices were measured.  Population growth didn’t cause more gold to circulate.

By removing the terms ‘demand’ and ‘supply’ from the AS/AD model, I hope to shake students out of their complacency, to show them how truly strange this model actually is.  It’s not a supply and demand model at all; it’s a model of how nominal shocks interact with sticky wages and prices to produce short run (but not long run) effects on real output.

P.S: Bill Wirtz has a new piece on global minimum corporate tax:

The conversation about global tax rules is tiresome because they deal in the age-old erroneous belief that companies actually pay taxes. Again, only three actors in a company can pay taxes: shareholders through reduced dividends, workers through receiving lower salaries, or consumers through paying higher prices. Nobody else can pay the tax; the building cannot, the carpets cannot. The fallacy of the corporate tax is, in its fundamental essence, not just that the negotiations between countries lead nowhere, but that the tax in itself is just another tax on consumers.