Songs for Spring 2026
I am a big fan of the AI song-creation app Suno. Currently, it will let you create 10 songs a day free, and will store and play back an unlimited number of your songs (as well as letting you discover songs by others).
Full disclosure: I learned about Suno from my daughter, who was associated with a venture capital company (Matrix Partners) that helped fund it. Although she is now leaving that company, she stands to receive a small fraction of its IPO or other exit value. I know I would love Suno even if it had no connection to my daughter.
Suno will take prompts to make a song about anything, including economics! If I can, I’ll try to make up a song or a few about the central idea(s) of each lecture. Listening to them is likely to help you remember key concepts.
You’ll remember the concepts even better if you use Suno or another song-creation app to make songs yourself in your own favorite genre. You can see the prompt I used, which has the key economic ideas in it.
The most annoying limitation on Suno is the limit on the number of characters I can use for a prompt. Running up against that limit, I often intentionally mispell words in order to use fewer characters. You will have to figure out how to deal with that limit if you make your own versions of these songs of economics.
Below are the songs, along with my commentary on them.
A. The Hockey Stick in Per Capita Income
Here my views are shaped by Brad DeLong’s book Slouching Towards Utopia: An Economic History of the Twentieth Century. By “Twentieth Century” in the subtitle, he actually means a “long 20th century” from 1870 to 2010. I do think that recently things have shifted, so ending things in 2010 isn’t bad though one could quibble. But why start the account in 1870? Because 1870 was about when the Second Industrial Revolution began. (Try prompting ChatGPT or other AI: “Tell me about the 2d Industrial Revolution.”) Brad DeLong argues that if the 1st Industrial Revolution (coal, iron, steam, textiles) hadn’t been followed by the 2d (electricity, cheap steel, fertilizers, dyes, explosives, pharmaceuticals, cars, expanded rail networks, mass production), that the “Malthusian Devil” of population growth outpacing resources for that 1st-Industrial-Revolution technology would have won again.
Here are 3 songs about the Malthusian Devil and the eventual escape from it:
I liked “Kings and Queens with Iron Fists” because Brad DeLong emphasizes that when humanity was poor institutions of force and fraud were the main strategy for a few people to get enough, at the expense of everyone else. Now, technology makes it so everyone could have enough, if we could only figure out how to do the social organization right—itself a very difficult problem that is actually the main theme of Slouching Towards Utopia.
All of these songs allude to the 1st Industrial Revolution, and “Dreams of Circuits” alludes to the Digital Revolution. But they give short shrift to the 2d Industrial Revolution. So I asked Suno to make a song directly about the 2d Industrial Revolution:
B. Models and Graphs
C. The Fisher Equation: Real Interest Rate = Nominal Interest Rate Minus Inflation
D. What the Fed Can and Can’t Do in the Short Run and in the Long Run
The main things I would add here if it had worked with the song are
(i) Control over interest rates in the short run also gives the Fed control over the output gap and a host of other things related to the output gap, such as the unemployment rate.
(ii) In the long run, the output gap is, on average, zero (this is beyond the Fed’s control).
(iii) The Fed controls very little in the long run, other than inflation and a few other things closely related to long-run inflation, such as the nominal interest rate.
(iv) As an exception to the above, the Fed probably has the power to ruin things in the long run by pushing the economy into hyperinflation or into secular stagnation by very bad policy. What the chorus of the song says and (i—iii) are what the Fed can and can’t do if it is trying to do good within at least a minimal understanding of how monetary policy works.
E. How Output Above the Natural Level Causes Higher Inflation and Output below the Natural Level of Inflation Causes Lower Inflation.
For this, I thought carefully about what I have written in research papers. Importantly, “There Is No Such Thing as Decreasing Returns to Scale,” so an upward-sloping supply curve does not come from a long-run U-shaped average cost curve. Rather, it comes from workers, factories and equipment in the factories being at least somewhat attached to particular firms. But also, tight labor markets and tight markets for rental spaces for business and other rentals shift a firm’s supply curve upward.
Firms “daydreaming” about price hikes or price cuts alludes to the “instantaneously optimal desired price” that a firm would wish to have right now if it were the only firm to have perfectly flexible prices. But, in fact, changing prices can be costly—at least if a firm thinks carefully about how to incorporate macroeconomic information into how it changes prices. (Decision-making costs can be substantial. Think of the level of education required to even be involved usefully in incorporating macroeconomic information into price-setting.) But eventually, firms do change prices, and macroeconomic information does percolate into those price changes. Exactly how that happens is a complex and contested area of macroeconomics. So the line “Dreams become what happens” refers to this complex process. A key aspect of this is that firms care not only about the instantaneously optimal desired price now, but the instantaneously optimal desired price in the future, since they might be sticking with the price they set for a while. If there is trend inflation, the instantaneously optimal desired price in the future will be higher than it is now, so trend inflation makes firms set prices higher when they do change their prices. The upshot is that inflation tends to sustain itself, unless output diverges from the natural level. Here are two songs, one for each direction.
(This was in danger of being very dry, I’m afraid, so I thought to put a little color on it by asking Suno to make Country songs about this. Hence the emphasis on farmers.)
F. Arrow Diagrams (“Directed Graphs”)
G. Open Economy Macro is International Finance
H. Divisia Indexes
I. Supply and Demand for the Monetary Base: How the Fed Has the Power to Determine the Safe Interest Rate
Take a look at Nick Timiraos’s article “Federal Reserve Posted Loss of $77.6 Billion in 2024” to see how the Fed handles operating losses due to its paying interest on reserves. Note that while this isn’t a serious problem, it does reduce seignorage. If you run into a paywall, note that you can get a free subscription to the Wall Street Journal if you are a student at the Unmiversity of Colorado Boulder:
As a student at the University of Colorado Boulder (CU Boulder), you have complimentary access to The Wall Street Journal (WSJ) through the University Libraries. To activate your free subscription, please follow these steps:
1. Visit the University Libraries’ WSJ Access Page: Navigate to the dedicated page provided by CU Boulder for WSJ access.
2. Activate Your Account: On the access page, follow the instructions to create or log in to your WSJ account using your CU Boulder credentials.
3. Start Reading: Once your account is activated, you can enjoy unlimited access to WSJ content online and via their mobile apps.
This initiative ensures that CU Boulder students, faculty, and staff stay informed with the latest news and analyses from The Wall Street Journal.
J. Why Housing is So Expensive
Many of the cities people would otherwise most like to live in have had strongly anti-construction policies. This is sometimes called “NIMBY”: Not In My Back Yard. In addition to directly raising the cost of getting land one is allowed to build houses on, the difficulty in getting permission to build makes it hard for companies to (a) get the immediate benefits of economies of scale and (b) get the technological learning that could have been gained individual firms each building many, many houses. Remarkably, there has been very little measured technological progress in building houses since World War II—a period during which we have gotten much, much better and making other things. See “Why Housing is So Expensive.” and (my renaming—it isn’t actually titled that) “Why Has Construction Productivity Stagnated? The Role of Land-Use Regulation—Reporting on the Work of D'Amico, Glaeser, Gyourko, Kerr & Ponzetto.” Here is the song:
K. The Overgrowth of Occupational Licensing (Another Reason to Pay Attention to State and Local Policy)
On occupational licensing, see “When the Government Says ‘You May Not Have a Job’”. Two definitions: “Occupational licensing” is when the government says “you may not do this unless you have gone through everything required to get a license. Certification” is when the government or some other organization says “you may not call yourself a [barber, Certified Professional CoActive Coach,BA,PhD…] unless you have met certain requirements”; those who are not certified can still offer to do the same job without certification, they simply can’t advertise a certification they haven’t earned.
L. How the Fed is Able to Raise Aggregate Demand
The prompt for this song was:
Make a song around this: 3 questions about the source of the Fed's power.1)How does a cut in the real rate raise aggregate demand? It raises investment, it raises capital flow&therefore net exports&it raises consumption thru the MPC in the Keynesian multiplier.2)How does a cut in in the nominal rate cut the real rate? "The Fisher equation: r = i minus pi and INFLATION is sticky because expectations are sticky" 3)How does the Fed cut the nominal rate? By cutting the interest rate on reserves
M. The Taylor Rule and the Triple-Strength Taylor Rule
Note: r* = r^n
Taylor rule: r = r* + .5 (pi - pi^target) + (.5/year) [ log(output) - log(output^n) ]
Equivalently: i = r* + pi + .5 (pi - pi^target) + (.5/year) [ log(output) - log(output^n) ]
Triple-Strength Taylor Rule: r = r* + 1.5 (pi - pi^target) + (1.5/year) [ log(output) - log(output^n) ]
Equivalently: i = r* + pi + 1.5 (pi - pi^target) + (1.5/year) [ log(output) - log(output^n) ]