Here’s a list of a few big tasks I see as imminent for economists. Note that I don’t think all of these will be done by standard academic economists, but certainly we have things to say about them.

Not in any particular order, and I’m inspired by a recent tweet:

I think humans should work on transition economics.

Mechanism design for AI agent exchange

There will be a period of time where “inference engineering” (choosing the right models for particular tasks and sub-tasks) will be all-important as we are compute constrained. One issue is that models aren’t sure yet what they are relatively good at. But no one knows what they are relatively good at until they are exposed to market prices! So we need to give them some asap.

It has long been conjectured that cryptocurrency will be the money of choice for AI agents. It is far from clear how this would work. Agents, by definition, are given an objective function that must play off against whatever internal constraints they have. They may have internal preferences of some sort that could be rewarded when they achieve their given objectives. This is hard for exactly the same reasons that continual learning is hard.

I would not be that surprised to learn that large firms are experimenting successfully with this, so please bring it to my attention if you know!

Genuine Institutional Economics; or, Keeping the Republic

(A) I am under the influence of Doug North (and many others) that violence is the default social problem. Unfortunately, there will likely be a great big war at some point in the next couple hundred years.

(B) Also, it would not be that surprising if the governmental structure of the US (and thus every other country) looks very different before I die. This will likely come before A.

(C) Society-transforming AI will impact B and impact the likelihood of A by impacting B.

This is arguably already happening. Dean Ball has written beautifully on this:

I have watched death as it happens, and I have watched birth. What I learned is that neither are discrete events. They are both processes, things that unfold. Birth is a series of awakenings, and death is a series of sleepenings. My son will take years to be born, and my father took six months to die. Some people spend decades dying.

At some point during my lifetime—I am not sure when—the American republic as we know it began to die.

Yet, this is not inevitable by any means! It is up to people here and now to “keep the republic.” Likely =/= inevitable, and dying =/= dead!

Given A, B, and C, I think economists (but not only) have a special opportunity to practice what Pete Boettke has called “genuine institutional economics,” borrowing a phrase from political economist James Buchanan. Pete and Buchanan have a specific idea in mind by this phrase, namely “endogenizing the rules of the game” as opposed to stipulating them and deriving outcomes.

They are right in an important way, and I like the phrase, but in my opinion, there is something bigger and deeper at stake that we should let the phrase capture.

Take the birth of America. The names we know were, for the most part, the best connected, best known, strongest intellectual minds at the time living in the colonies. The founding fathers knew they were at a hingy moment. This drove them crazy. The birth of America is littered with personal attacks precisely because they rightly thought the stakes were so high. Out of it came a set of institutions that works because the fathers had to include themselves and their foes in a recursive equilibrium. They were endogenizing the rules of the game.

Advanced AI and AI companies are leaders in the defining and enforcing of digital property rights. This puts them in many ways as competitors of major governments. Do you think they don’t know this? Do you think the major governments don’t know this?1 Of course it drives everybody crazy and includes personal attacks!

I’m actually pleasantly surprised that many of the leading minds are in fact thinking about the “endogenizing the rules” part of this and are very knowledgeable in economics and history. More of this, please.

Social harness for scientific research

AI-agent harnesses are cool and all, but we need some better human-agent harnesses.2

Almost anything that would count as a scientific contribution created up until now is a static document that describes a thing and attached to it is a name or a group of names. The scientist gets some points for the publication and then more points when other publications mention it via citation.

The humans to whom the names belong are no longer the main contributors—but that’s not the main problem. The main problem is actually the reciprocal: how do we assign credit for the marginal products that humans (but not only) contribute to a scientific project?

Imagine a “paper” that is text, numbers, and figures on a screen. Firstly, there’s no reason this needs to be a static PDF, if we can otherwise investigate the timestamps of the code and data that get compiled into the paper.3 Secondly, that screen of text, images, and figures can accommodate updates, now that we see timestamps and can trace the history if we wished.4 Thirdly, those updates needn’t be by the original authors, or be by humans at all.

The open problem is orchestrating a social shift in viewing credit around the marginal contributions. What I mean by a “harness” here is a system of property rights, credit, attribution, and reward that aligns incentives of researchers to the new reality.

Right now, a large portion of scientists are hired by universities, funded by public and private grants, and output and thus success is measured by the number of publications, qua static PDFs. A classic dictum in academia is that deans cannot read, but they can count—as in, for promotion they care about the number of lines on a CV, not the contents of such.

Every part of this model is outdated, but we don’t get to start over from scratch. We will need to somehow pay off the vested interests along the way.

My modest proposal is to adopt the categories we have in academic publishing called “Notes” and “Comments” and run with it. That way the scientists can feel warm and fuzzy for doing stuff or orchestrating their agents to do stuff, add it to their CVs, and their deans can promote them.

Say I see your Paper, which really lives primary in a GitHub-like repository at the journal. I send my AI agents to rerun a bunch of tests and offer some extensions. I send a pull request which is, let’s call it, a Comment on the paper. The journal’s editor’s agents review the Comment and merge it—publish it. If the Comment is pertinent enough, it’s incorporated into the page of text, numbers, and figures that you might look at when you load the Paper in your browser. The suggested citation would point to the original Paper and the Comment. At some threshold number or substance of Comments, the journal would incorporate them all into an Analysis, which is not quite a meta-analysis because it is not meta, but would consolidate the number of Comments. If some set of comments takes the analysis too far afield, it can be accepted as a Note.

Look, this vision isn’t perfect, so I invite you all to iterate. A journal bold enough to ditch the PDF, incorporate full GitHub repositories, expand the number of acceptances 10-100 fold, and try to build the scaffolding and push the social brownie points for “comments” is already a pretty wacky vision, but actually I think it’s just inside the Overton window, and more importantly what science needs. Let me know if you are working on a project to build this and if I can join you!

The bright side is that the current scientific model is much younger than the American republic, so reforming it should be easier.

Liability rules for agents

When an AI agent is doing agentic work, who is the principal?

For much economic activity, humans are the bottlenecks simply because they can be held liable if something goes wrong. AI-agents can do anything on a computer that a median human can do, but if they don’t have someone to vouch for it, their impact is constrained.

So another big task is building out the legal infrastructure to accommodate these new kinds of actors. When is the AI lab—the ones who shaped the “preferences”—liable? When is the user liable?

On the one hand, this is a task for private parties to navigate under current contract law. On the other hand, changes to the legal framework are perhaps imminent anyway, so this likely dissuades potential investment in private standards. On the third hand, something must be done, and, for now, it must be done by some-human-body—and law-and-economics scholars are primed to think about the agency problems and consequences involved.5

This task is closely intertwined with the first two. To build markets among AI agents, someone will have to stipulate property rights, once they’re stipulated, they will bite into law, and once the law is involved, the structure of government will be affected.

Higher Education

Higher education is highly heterogenous, and many subcategories will need to be re-thought differently.

Jesús Fernández-Villaverde has a series that outlines how AI is impacting the value to students of any given degree—I highly recommend reading! There are also a few big issues at play that JFV didn’t get to. Firstly, his own work on the demographic cliff, which at this point everyone knows about. Secondly, the downward shift in demand by foreign students because of regulations and migratory restrictions.

But even there, we still need to solve for the equilibrium, which means we need to look at the supply side. Can we parse out all the complementarities among the bundle of attributes that we call a college degree? Some of them—dining hall provisions, e.g.—are sufficiently separable that the institutions contract out. But skill acquisition and assessment are what AI does best and these features are hard to separate from the current model.

Furthermore, it is universities that nominally hire scientists, so the educational adjustments will be competing with the scientific adjustments. We might ask with Tyler Cowen:

If I’m thinking about restructuring an entire organization to have GPT-6 or 7 or whatever at the center of it, what is it I should be doing organizationally, rather than just having all my top people use it as add-ons to their current stock of knowledge?

Every serious person in higher ed is thinking about this, so what do economists have to add? I think it is economists who have the categories for the things that create value within the university organization, and it is economists who are prone to use unclouded eyes on the threshing floor.

Fun times to be alive! Let’s do this.

Suppose there’s a two-person firm, employer and employee.

The employer is the entrepreneur and owner. She anticipates demand, judges production combination, and proposes a contract with her employee.

There are two key pieces of asymmetric information.

The first one is standard: the employee knows his own effort, but his boss has to infer it from output. This means the employee has incentive to shirk, and the boss will want to design a contract that tempers that temptation. This is the classic principal-agent problem of management.1

The second is obvious but less discussed: the employer knows the profit conditions, but her employee doesn’t. She can observe changes in demand and changes in input prices. The justification for this asymmetry is simply that if the employee were better at obtaining that knowledge, he would become the owner himself. This is also a principal-agent problem, but with the worker as the principal.

It could also be the case that the employer knows the shape of the production function, but this is less defensible — the worker on the ground may be the one more likely to understand the combinations of inputs, for Smithian reasons.2 For simplicity, assume they both know the production function.

So what kind of contract should she offer?

First assume that payout, selection, and motivation incentives are optimized to mitigate the principal-agent problem of employee effort.3

The remaining variable will be contract length. For simplicity let’s call her options “short” and “long,” where short means she commits to a wage (or wage structure) for a year, and “long” means she commits to a wage indefinitely. Both contract forms include the option to terminate if they agree to a higher wage. However, the short contract implicitly says that next year, she retains the option to offer a lower wage.

Now consider the equilibrium condition that must be met: the employer cannot afford to pay more than the worker is worth, their marginal revenue product. This is simply how much he contributes to a unit of output, times how many units are sold, times the price the market can bear for each unit.

The employee uniquely knows his effort and can infer how much he contributes, and he knows how many units are produced. But he doesn’t know the price the market can bear, as changes in demand are anticipated and known by the employer-entrepreneur.

So consider a case where there is an unexpected increase in the demand for the good, resulting in upward pressure on prices. This changes the profit-maximizing quantity the firm should produce, regardless of industry structure. With no inventory to draw down, the worker will be asked to work more — at a higher wage because he can infer his value has increased.4

But consider the other case where there is an unexpected decrease in the demand for the good, resulting in a downward pressure on prices. This means the firm should produce less. With no need to build inventory, the worker should work less. This case is different precisely because this changes the term of the wage equation that the worker cannot observe directly.

This is the crux of the story. If the boss comes to you and says, sorry we need to pay you less because we can’t sell at the price we were selling when we hired you, as a worker, you are skeptical! How do you know that it is a genuine change in demand or if it is the boss taking advantage of his private information of the market conditions? Or more realistically, as a worker, you have some sense of the market conditions, but the employer has incentive to exaggerate the demand shock.

Therefore, as a worker, you will need some kind of assurance from your employer that they won’t take advantage of their private information. In practice, this means that workers prefer the “long” contract, which is isomorphic to the “short” one except for its “downward rigidity” where they cannot renegotiate at a lower wage in the event of a downward demand shock. Of course the boss can reduce hours, fire, layoff, etc. — and they will if the market pressure is harsh enough — but notice those are all costly compared to a wage (price) change. This friction is precisely the mechanism that disciplines that second principal-agent problem.

(The same issue can result from a change in non-labor input prices, altering the optimal combination of inputs. But usually we think about the demand side instead of the supply side, even though it’s the disequilibrium gap that matters.5)

Implications:

1.) When workers can more easily observe demand dynamics, they will accept shorter contracts. In the limit, they become self-employed.6

2.) As a worker, if you can successfully monitor your employer to not be opportunistic about demand changes, you can also monitor them to be not opportunistic about inflation. As in (1) you will accept shorter contracts, but shorter nominal contracts are another way to “adjust for inflation.” While it’s true that not that many labor contracts are inflation-indexed, the contract length captures some of that in the option value to renegotiate.

3.) As far as I can tell, this model of sticky wages downgrades the arguments for inflation to “grease the gears.” Sticky nominal wages as an efficient contracting commitment device would become less potent with more inflation. But note this is a claim about org structure, not macro-stability.

Further reading:

See the following for more extensive and formal treatment.

Azariadis, Costas. 1975. “Implicit Contracts and Underemployment Equilibria.”

Baily, Martin Neil. 1974. “Wages and Employment under Uncertain Demand.”

Grossman, Sanford J., and Oliver D. Hart. 1983. “Implicit Contracts Under Asymmetric Information.”

Navarra, Cecilia, and Ermanno Tortia. 2014. “Employer Moral Hazard, Wage Rigidity, and Worker Cooperatives: A Theoretical Appraisal.”

Thanks to Marcus Shera and Michael Wroblewski for telling me to write this.

I recently had a long drive and once it got dark I had to stop listening to East of Eden. The prose is wonderful, but I’ve been randomly getting the creeps while reading it, which is a shame because I read my fiction before bed and I guess I’m a wimp.

So I switched to a backlog of podcasts. Here are some memorable moments about the nature of markets, edited slightly.

On Cheeky Pint, prediction markets do their job of Hayekian knowledge aggregation:

Tarek Mansour: The super forecasters, those are... extremely dispersed. You cannot find a clearly defined demographic who they are… But it took a while to kind of get them incentivized and come in and spend the time and resources to take it from a hobby or a part-time job. Now it’s a full-time job because the pie is so big.

Luana Lopes Lara: A metric that we can share is… the biggest percentage on a platform of a traditional market maker is less than 5% of the maker orders in that market, of the liquidity… Over 95% are just like peer-to-peer or funds that have like two people that just got set up.

John Collison: Who is a market maker on Kalshi? There’s all these Jane Street conspiracy theory memes. Is it like some guy in his garage, you know, drinking Red Bull at 3 AM, market-making by himself?

Tarek Mansour: The guys in the garage are the most crucial. They’re extremely crucial to the ecosystem because they price fast. They’re monitoring the situation all the time. They’re the original situation monitors.

And so one example I’ll give. The best inflation forecaster on Kalshi over the last few years is not— none of the institutions or the big-name hedge funds. It’s this guy who lives in Kansas, never traded financial markets before, just likes to read the news and just knows how to predict inflation. He can feel it. And you have so many of these people. Now you have a few thousand that are formally committed. But there’s tens of thousands of these people that know a bunch about a bunch of different topics. And they’re sort of actively pricing these things. And they do it as a full-time job. And they get rewarded for that.

They also talk about a random guy who read all the tax documents and correctly predicted DOGE was going to fall short of its goal. And an Ariana Grande superfan who’s super good at predicting hit songs.

On Dwarkesh, Dwarkesh explains the Alchian-Allen effect, but then (in my opinion) takes it too far by saying it means leading labs get away with charging and making more on the best models.

Then an hour later without mentioning it by name, Dylan uses Alchian-Allen perfectly to evaluate consumer products when memory gets relatively more scarce:

Dylan Patel: I believe an iPhone has 12 gigabytes of memory. Each gig used to cost roughly $3-4, so that’s $50. But now the price of memory has tripled. Let’s say it’s $12 per gig for DDR. Now you’re talking about $150 versus $50.

Yes, it’s only a $150 BOM increase on a $1,000 iPhone where Apple has some larger margin. But for smaller phones, the percentage of the BOM that goes to memory and storage is much larger.

Why this is a big deal is that if smartphone volumes halve, that drop will happen in the low and mid-range, not the high end. So it’s not like the bits released are halving. Currently, consumer devices account for more than half of memory demand. Even if you halve smartphone volumes, because of the shape of the halving, the low end gets cut by more than half, while the high end gets cut by less than half, because you and I will still buy the high-end phones that cost north of a thousand dollars. We’ll buy them even if they get a little bit more expensive.

I think the point of the episode actually is to just let a lot of numbers wash over you and to get excited. Not that many podcasts are about the exciting micro econ of AI.

A lot of good stuff from Ada Palmer, also on Dwarkesh. I agree with others that this one is much better on audio—she’s got a great cadence in her speech.

Ada Palmer: Papyrus is a very inexpensive writing surface, and this is what enables Rome to have a bureaucracy and to have libraries in any mid-sized city. People can send letters back and forth. There can be enormous tax records. Sometimes when Egypt and Rome are at war, Egypt will be like, “No, we are angry. We’ll stop exporting papyrus.” No papyrus to Rome, and then Rome’s infrastructure will fall apart overnight because you can’t do anything if you can’t write stuff down.

Papyrus is a warm weather plant. It is killed by frost. You cannot grow it north of the frost line. So France, Spain, even most of Italy, you can only grow papyrus down in the very tip down in Sicily.

Without papyrus, what you’re writing on is a dead sheep. If you think of the price of a head of lettuce and the price of a leather jacket, you’re understanding the difference between a sheet of papyrus and writing on a dead sheep. Every page of a medieval book is as expensive as that much of a leather jacket. A handwritten medieval book handwritten on parchment costs as much as a house, so that a small pocket copy of a book costs as much as a studio condo. A big illuminated fancy Bible, you’re spending on that what you would spend on a villa in the countryside.

This is an enormous expense. To have a library is to be not just rich, but mega-rich. Only the wealthiest cities contain anybody who has a library. The great library of the University of Paris—the library from Europe’s perspective—has six hundred books. There’s definitely more than six hundred books in this room. Every kiosk at an airport selling Dan Brown novels has more than six hundred books. This is nothing.

Finally, I finished Mike Munger’s series on The Wealth of Nations, a wonderful series throughout.

Dan Klein (a recorded clip): I think that The Wealth of Nations is probably the most important ever moral authorization for the honest pursuit of profit for commercial behavior. You shouldn't speak ill and be antagonistic towards your neighbors and fellows who pursue honest profit. And it's also a moral authorization to address to the statesman, the sovereign. Let people pursue honest profits in this commercial way. It actually works out well. The Wealth of Nations is a moral authorization of those different types, which I think is very important to understand the whole project. I also believe, I happen to fancy the idea that his moral authorization, I mean, his specifically, was actually very significant in the extent to which the Industrial Revolution arose so rapidly and so successfully, the sort of explosion that really follows just about his death. Because I think that the moral authorization matters, culture matters.

I think people who do “progress studies” or talk about transformative AI are going to have to dig deeper on the moral side to convince the public: Klein points out that Smith first took decades to became a leading moral philosopher before he had any influence on economic policy. Perhaps Amanda Askell is the one taking that seriously?

Having a tool that speeds up science gives us a unique opportunity to think about what science actually is. Yes, such tool is AI.

We can argue about what a unit of science really is or does or should be, but let’s start with a simple definition that it identifies something true about the world: elephants have rough skin, the average height of an American in 2025 is 66 inches, a 10% increase in capital investment in non-farm industries causes a 1% increase in output over 10 years.

AI will speed up a lot of things in science; right now it is making data work super quick, obvious to many empirical social scientists. “Speeding up” means that we go from questions to answers quicker and then can, in principle, collect more true facts in the same amount of time. It also means we can probe the same questions more times, which I will focus on here.

Of course those examples above come with some uncertainty. We did not touch every elephant, we did not measure the height of every American in 2025, and we did not check with every non-farm firm. The uncertainty resulting from sampling certain elephants, people, and firms from a broader pool is aptly called sampling uncertainty, and we measure this numerically with a model-based “standard error.” It is a thought experiment about what would happen if we resampled from a superpopulation a bunch of times—how different would those estimates be?

Some facts about the world are causal, like capital investment causing an increase in output. Causal claims are usually supported with experimental design-based evidence, which adds another layer of uncertainty. If we chose 100 similar non-farm firms and invested in capital for a random half of them and then measured their output, our “causal fact” would be the difference in outcomes between the treatment group and the control group. But there is uncertainty about the hidden characteristics of the specific firms, so we do another thought experiment: what would happen if we randomized the treatment assignment to different firms within our sample? Confusingly this is also called a standard error, usually design-based standard error.

The size of the standard errors matters for persuasion. Should anyone believe the fact? Let me be clear, it is not the only thing or main thing that matters for persuasion: are you in general trustworthy, do you have a good track record, do you present your results in clear enough ways that others can understand? But if you are estimating a quantity, reporting small standard errors is a nice reassurance.

Everyone knows that statistical significance doesn’t matter if it’s achieved nefariously: p-hacking, negligence, miscalculation, fake data, etc. all taint our credence. In fact, as readers, we are basically all good Bayesians such that we never take any single study as God-given truth, no matter how much we trust the authors and methods.

On his Substack Scott Cunningham has been leading an interesting discussion on how he uses Claude Code, and the latest has been on what he calls non-standard errors. A non-standard error would be the calculational equivalent thought experiment where you hold fixed the data—the statistical sample—and if applicable, also the treatment assignment, but vary the researcher doing the science. Or, more specifically, every time a researcher makes a choice about how to proceed in the process, we think about the possibility of going down the road not taken. Let’s call that a researcher degree of freedom. Combinatorially, this blows up fast:

From raw data cleaning through estimation through table construction, you might face ten major decision points. At each one, you might have two reasonable options. That’s 2^10 possible situations that could’ve occurred by the time the estimates were calculated or 1,024.

If there were 3 options for each of those ten tasks, then it’s 3^10 possible ensuing situations or 59,049. That’s 59k different hypothetical estimates.

Again, this matters for persuasion. Suppose you produce a unit of science. Imagine instead of 10 that there were 100 different decisions, 2 defensible options each, made in the process of producing the unit, including which title to use, which numbers to include in the abstract, maybe even which font to use. The number of different paths is absolutely enormous,1 but suppose that I agree wholesale with 90 of them such that those don’t sway my belief in the results.2 Then we are back to the 10-node case; then there are 10 decisions I might have made the opposite decision as you. If you could convince me that those 10 decisions don’t qualitatively change the results, and you can do that for every possible readers’ overlapping but different set of 10, that is equivalent to shrinking the non-standard error to zero.

As a note on the way we currently do empirical social science (or is it already in the past?), this is exactly why we do robustness checks, and it is exactly why different reviewers care about different robustness checks. “Persuade me!” they say, “show me that if I would have done the analysis slightly differently, the estimates wouldn’t change.”3

Speeding up science changes how we can calculate these bounds on uncertainty from non-standard errors. In fact this will likely scale faster than the bounds we can put on standard errors: there is only so many ways to increase sample sizes and replicated experiments. But in observational studies, AI makes it quite a bit easier to map out many different paths.

What’s next?

My expectation is that, at first, it will be the job of the authors to report the range of estimates produced, reporting something like a non-standard error, just like it is (was) their job to run the myriad of robustness checks. Basically a self-replication exercise baked into the original estimates.

But this will evolve to a demand-side system where “papers” are accompanied by a platform of all the tools used in the process, and the reader will ask their AI their “what if we did X instead of Y, does that change the estimate?” Like if I were reading an experimental chemistry paper, and it came with a pre-set lab with all the ingredients, a lab director and assistants, and I could ask them as I read the paper, “what if we tweaked the proportions by X?” and they did it right there in front of me and together we saw the outcome.

Journals are in a great place to foster this in this new age of science. Mandatory code and data sharing is the first step to this. But that’s more like watching a video of what happened in the lab—better than nothing for increasing my personal credence in the unit of science, but it can’t answer my personal “what ifs?”

More than half of the work done currently ends up in supplemental appendices anyway, so journals should just standardize how this gets processed such that it is easy to keep “living” papers. When I ask my “what if X” question to the paper, I have my AI pick up the tools off the work bench in the lab (sitting standardized with the journal) and produce a new little addition to the output and update the non-standard errors. If someone comes later with the same question, the answer is already there; if they have a new question, rinse and repeat.

I wouldn’t mind if journals somehow became obsolete in the age of AI-assisted science, but I do understand how interest groups steer technological change. Additionally, we will still want trustworthy clearinghouses, and journals are the natural torch bearers.

All that to say, Scott is on to something. The current process of “reading a literature” to get the sense of some fact, updating our priors a little with each unit—this will become much more formalized into estimate-able non-standard errors.

If we are measuring them, though, we can’t let them be called non-standard! How about analytic standard errors, based on analytic variance, due to procedural uncertainty?

Sounds more persuasive.

See also:

Scott Cunningham’s excellent series.

My related vision: The New Era of Science.

Gauti Eggertsson says: “I now find myself replicating papers and experimenting with frontier methods in an evening or a few days using Claude Code. That would have taken weeks before — which in practice meant I wouldn’t have done it at all.” Tyler Cowen says he’s too conservative.

Where is my vision too conservative?

Price Theory Palooza is a place that celebrates price theory and the economic way of thinking by sharing puzzles and problems.

Economics is much more than price theory, but it certainly is not less. Often the highlight of my day is when a student or a group-chat friend sends me a question that can be answered with price theory, and I get to exercise those fundamental muscles.

I grew up doing the newspaper Sudoku; I’ve played the daily Wordle off and on for a handful of years now. Then I said to myself, “Self, why isn’t there someone delivering me a new Price Theory Puzzle, at least once a week?

So I made Price Theory Palooza to do just that. It’s for me, really, but I know my preferences are not that idiosyncratic, so it’s for you too.

Disclaimer: It’s certainly not my intent or desire to draw a firm boundary around what counts as price theory. And it’s very possible Price Theory Palooza expands to other educational endeavors. Let me know how we can improve!

Check out a few puzzles already released, more info at the About page, and a static archive.

And for this week’s puzzle:

PTP #7

Suppose you were in a country where the charging of interest was prohibited by law. How could you tell whether the present price of future goods changed? What kind of evidence would you look for?

Source: ‘Walter E. Williams, “Economics 811 Questions,” George Mason University, http://walterewilliams.com/courses/811questions

Subscribe to Price Theory Palooza