In our previous installment, we talked about the idea of an economic model and introduced a large model for an economy. We got some good discussion out of it and ordered our thinking on how a modern economy works. But just as important, we established the idea of a model, which is to throw out lots of information in order to find the underlying patterns which might otherwise be obscured.
In this diary, I'd like to talk about how an economist sees time and introduce the model of the short run vs. the long run. This model shows up as an aspect of a lot of later models, and I'll discuss one later model that has a lot of relevance to our life now -- the Keynesian Multiplier. The Keynesian Multiplier is a discussion of how the government can help us out of --or into -- recessions, and it has gotten a lot of empirical (real world) support lately.
To an economist, time basically comes in two flavors: the short run and the long run.* This is one of those sets of ideas that sounds simple but is actually very powerful. It solves some very confusing questions in how people organize themselves. The short run vs. long run dichotomy also produces some of the most effective ways to lie with economic models, so it is vital to understand it.
The short run is defined as the period of time in which things that are easy and quick to change can be changed, but things that are difficult and costly to change cannot.
The long run is defined as the period in which everything which can change has, and the full costs of each option are felt.
I'd like to illustrate the difference between the short run and the long run using a restaurant. In the short run, a restaurant can extend its hours, put a few more tables onto the sidewalk, or hire some more waitstaff and kitchen staff to keep customers moving. In the long run, it can remodel, change locations, or build an addition for its clientele.
In the short run, if a school has a blip decreasing the number of students in that area, it can lay a few teachers off and run a couple fewer buses. In the long run, it can move to a smaller facility, shutter (or rent out) a section of its building, or consolidate districts.
In short, the short run consists of actions which can be quite costly but can be implemented quickly. The long run consists of any change you can imagine. The best example of how long the long run is comes not from the Depression (which is sad) but from the Apollo Program, which turned us from a species which had not gone to the moon to a species which had -- an enormous difference -- over the course of a decade.
There are several reasons why this concept is powerful, but perhaps the most useful aspect of it is the fact that the short run is a different duration in different industries. It is perfectly possible to renovate a space and open an entirely new restaurant in three months. It is not possible for an automobile company to open an entirely new assembly line in less than a couple of years. That's the power of the short run and long run model; it lets us group problems which are similar (the opening of new firms or new branches of firms) without being hung up on the differences (the amount of time it takes to change things in different industries).
This grouping lets us understand a few patterns:
1) Short run changes involve labor, long-run changes involve capital. You'll note that in all my discussions above, labor movements were the easiest to make. In the short run, it's quicker to staff up or down. In the long run, more profound changes need to be made.
2) Some prices can change in the short run, but by and large relative prices stay pretty constant. In the long run, relative prices can change dramatically as demand and supply for goods, labor, and capital change dramatically.
3) Sometimes, the long run can be pretty darn long. A decade for the moon shot, for example. Or thirteen years for the Great Depression. Other times, it can be really short. This variation explains why some industries (and economies) seem to bounce back from problems, while others take time and involve huge transition costs.
Ok, so now we have some new intellectual tools that we can apply to problems in front of us. The first one I'd like to apply it to is the Keynesian Multiplier.
The idea of the Keynesian Multiplier is that government spending can be a more powerful effect on the economy than just the value of the spending itself, because people who receive it then buy other things, recirculating the stimulus. However, this effect only exists in the short run, because in the long run, people basically go back to work and can't be pushed without heroic effort. Besides, eventually you have to pay off at least some of the debt, and that creates the exact opposite problem.
So: how big is the Keynesian stimulus? Well, that depends on your Marginal Propensity to Consume (MPC). Let's take apart that phrase. Marginal means a small change. Propensity is your likelihood of doing something. And of course consumption is using something up. So the Marginal Propensity to Consume is the amount that you will consume, if you get a little more money. It is between zero and one.
So, I get a dollar of stimulus. That's a dollar. Then I spend my MPC * 1 dollar and someone else gets that. That's 1 + MPC.** Then the schmoe who gets my dollar spends their MPC of that. So now we're at 1 + MPC + MPC2.*** You can see where this is going, eventually, and mathematically, it looks like this:
1 + MPC + MPC2 + ... + MPC∞
Due to a quirk of mathematics, that comes out to:
1****
-------------
1 - MPC
Those of us who remember our algebra well note that as MPC goes up, the denominator gets smaller. And as the denominator gets smaller, the total gets larger. This brings us to the fundamental truth about the Keynesian multiplier:
The more people who receive stimulus consume, the larger the Keynesian multiplier is.
And that is why, if you want to goose the economy in the short run, you should find a group of people who are likely to spend most of their new money. That is, you want persons who have immediate needs, such as unemployed people and poor people. What you don't want is people who will just sock the money away, like wealthy people and monks who have no need of material possessions. That's why a tax cut toward the wealthy is the very worst kind of stimulus, and directly hiring the unemployed is the very best.
Well, I'm out of space and I'm sure there are plenty of questions. So let's wrap this up for now. To sum up, today we learned about how economists see time, in terms of the short run and the long run, and we learned about the Keynesian Multiplier, which is in the news a lot these days. Tune in next week, when we discuss the discount rate and negotiation, or an economist's understanding of power.
*Economists also talk about the "medium run." This is, unsurprisingly, the period of time which shows characteristics of the short run and the long run both. It refers to a transitional period.
**Oh ho ho, you say. What if this isn't a closed economy and some of my money goes overseas? Yeah, that's a thing. It's why the Euro countries had to agree to do stimulus together. The good news is, the way the math works out, it doesn't affect our conclusion that stimulus depends deeply on the MPC.
***What if my MPC is different from the MPC of the person who is spending the dollar? Good question. Well, in aggregate, the people receiving the money from the directly stimulated people should be roughly the same MPC as the national average. But the math on that one is a bit complicated, and we'll set it aside to be examined in a comment below once I have a chance to figure out how to explain it well.
****What about tax cuts? Well, they work the same, except that there is no initial stimulus; the only stimulus comes from the spending of the people who got the cut. So instead of 1, the numerator is only MPC.
Update: Thanks again for the community spotlight! I'm deeply flattered.