The twenty-page essay exam

Imagine giving all professional economists (and other academics) an essay test.  Determine their area of expertise, and then ask them to write a twenty-page essay on one of the most basic questions in that field.  So it might be “Why did China do so well?”  Or “what are the determinants of economic growth?”  Or “What causes business cycles?”

Some would be more specific, such as “What makes nominal prices sticky?”, or “Why are the special features of platform competition important?”  How about “How can we encourage hospitals to compete more effectively?”

They can use some numbers, but mostly they should write out the best answers they can.

Then grade the exams.

At which universities would professors do the best?  In which fields would the researchers do the best, recognizing that some face more difficult problems than do others?  In which countries?

How much should our profession be focused on being able to write good answers to such questions?

I am indebted to Chris Blattman and Arnold Kling for useful exchanges related to this topic.


Fun question!

By university, I do think there's an easy answer - it's whichever universities have their professors teach more introductory courses. That exercise forces profs to be able to go back to the basics and explain them well, whereas in universities where the professors can spend 100% of their time on research, they never have to do stuff like that. So I think you can estimate the answer to the question of "at which universities would professors do the best" by ranking universities by the proportion of time that professors there spend on teaching.

Great answer.

Another possibility is the universities employing most professors who had spent their undergraduate years writing two or three essays a week. (I'm allowing for the likelihood that middle-aged gasbags are unlikely to be as terse as sharp, young undergraduates.) Probably Oxford and Cambridge, then.

Maybe, maybe not. It is probably logical that an educator would think that a test would tell you how smart or effective someone is. But it is likely that all of these people would write good comprehensive essays and yet all disagree. So now how do yoou grade them on the quality of their written words or the content of their argument? Either way would you reward the best writer and possibly ignore the most competent person in their field? It isn't that simple.

Stupid question unless you know who is grading the exams.

BTW I choose Krugman to grade the exams, because he would be blunt, or rude, enough to call it a stupid question as stated,

As many teachers and professors have noted, "life is not an exam". Good answers to such an exam, while nice, would not be the same as identifying who are the good economists.

But that's not to say that the exam scores would be uncorrelated with the economist's knowledge, ability to write, and ability to analyze and synthesize. Don't most (all?) econ departments have questions similar to these (albeit typically narrower) on their general examinations?

MIT's and Stanford's don't seem to be online in an obvious place but Harvard's recent general exam questions are here:

The essay questions on the 2016 macro exam include ones such as this:

"Can you list four distinct reasons why global real interest rates on advanced country government bonds might be at exceptionally low levels? Just a sentence or two on each reason can be enough for full credit."

Real interest rates, from 2008 to 2014 in the US and to 2017 in Europe were at exceptionally high levels compared to those give by, say, the Taylor Rule. Hence the depressionary nature of the Great Recession. And hence the depressionary nature of depressions, as opposed to recessions. And this in Rogoff's section, for goodness sake.

Looks like a fun exam up at Harvard, though.

I have always thought that a competent academic ought to be able to explain what they do in under five minutes and in simple clear English.

The problem is that most of the people I know who can do this are Engineers or otherwise in the hard sciences. The Humanities fields do not seem able to produce people capable of this. Not the five minutes. Not the simple clear English.

I have long since concluded it is because they know their fields are fraudulent and if expressed simply and clearly everyone would notice. You can only claim carbon fiber is a sexist and racist construct if you use very complex language indeed.

So I would say twenty pages is too much. How about two pages? Better yet, a two minute video.

I'm reminded of a Mark Twain quote (or was it Samuel Clemens, I get those two mixed up): "I didn't have time to write a short letter, so I wrote a long one instead". 10 pages long would be twice as hard and twice as clear.

I now want to study clarity vs. conciseness, see if I can create some sort of formula.

Brevity is the soul of covfefe.

Brexit is the soul of covfefe... and it’s a good thing too.


But the underlying idea is a good one!

So Much is correct. If you need 20 pages to explain what causes business cycles, you have no idea what causes business cycles. You are throwing crap at the wall and then lunging yourself at any possible concept to cause it to stick. And then hoping that after 20 pages, some part of it convinces somebody. Or at least kills them through attrition.

Shame on any professor who allows this, by the way. If you ask this kind of question to your students, you are admitting that you do not know the answer, either. And also that the entire profession is all about blabbering endlessly, in a convincing fashion.


I’m in History and years ago when I was stumped with respect to my dissertation, I was fortunate enough to have an aunt (they’re not gentlemen, they’re often better!) ask me to explain my thesis in 50 words or less, and clearly to boot. That helped me immensely, and the aunt was happy too.

Out of curiosity, closer to the work Sarpi did on the Council of Trent or McMeekin's work on the Russian causes of WWI?

I tried my best for equipoise in that question

"The Arabs have an expression for trenchant prose: no skill to understand it, mastery to write it." Taleb

That's not how economic research works and probably not how it should work. It's a fun exercise, but given that it's a Cowlerian principle that everything is about status and that you limited this to "professional economists (and other academics)": I don't think we should assign status based on something like that.

This is a question that I expect economists at Think Tanks, Bloggers and Journalists to do well. It requires wide knowledge and good writing skills. Economic research needs curious people that are able to concentrate on one topic over long time periods.

The proposal does not exclude research as it is currently practice.

It seems like all the work here is being done by the process of grading - who is doing it and what their criteria are.

I agree. In particular, the grading process determines who will do best.

Just give all the professors a copy of Elements of Style and instruct them to write like that. Why do economists write in a language only known to them? To prove to their betters that they can? Clarity in communication is the sina qua non (the irony is intentional) of understanding and consensus, not to mention order and stability. Okay, I can appreciate that academic papers aren't users' manuals (although I can't recall a user's manual that is intelligible). And here at MR readers enjoy interpreting Cowen's Straussian code. But can't anyone write anymore? To be fair, a writer must know his audience, and today's audience has neither the patience nor the capacity to digest well-written prose. What the audience wants are images, pictures and graphs that supposedly illuminate but more often than not merely deceive, or at best distract and entertain. What is the intended audience for academic papers, and is it the audience that is to blame?

I still remember reading a users manual for VMS, the operating system for DEC VAX computers, circa 1988. A model of clarity and fitness for purpose.

I’m reading (finally) Kissenger’s book Diplomacy now, and was pleasantly surprised by the clarity of the writing. It’s a pleasure to read.

"Why do economists write in a language only known to them? To prove to their betters that they can? " Says the lawyer? Really?

The essay test is an intriguing idea. For bonus points, articulate your topic to an average person.

Which graders will do best? Loop question.

Beyond clarity (clarity being my first comment), I suspect Cowen may be addressing more fundamental questions: what is the consensus American view and is the consensus American view the best view. As to the first question, I'd say there is no consensus American view; indeed, what we have learned over the past year or two is that America's economic consensus is adrift. As to the second question, I' say that there is considerable doubt about whether the consensus American view will work in today's global economy; indeed, what we have learned over the past year or two is that the consensus American view has long since had its past due date. Cowen framed his blog post as the ability to write good answers, but I suspect what he means is the ability to write good answers: what is the good answer. Read this: Don't be deceived by the title. The essay it isn't so much about the issue of free trade, but the role of the state in the economy. If the answer for a robust economy can be found in the China model, I fear that we can't get there from here. By that I mean China moved from a government controlled economy to a more open economy but with an active government role. I'm not sure a country, this one anyway, can go from an open economy with a very limited government role to a less open economy with a more active government role. We can't write good answers unless we are willing to ask good questions.

TC: "What makes nominal prices sticky?" - this is known as a leading question in law, and not allowed unless your own sympathetic lawyer asks it. Replace "What makes" with "Why or why not" and the question is OK.

This reminds me of the Nasrudin tale where he is accused of blasphemy by his learned enemies. In the courtroom he asks that his accusers be given pen and paper and answer the questions, "what is bread?" Naturally for each person the definition varies: wheat+flour+yeast, something on the outside of sandwiches, and so on. Nasrudin confronts the judge to say that his accusers can't agree on something they eat every day but they can somehow all agree that he is blasphemous? Of course he is acquitted.

So I'm not sure what you think the results of the papers will be, but if you hope them to be graded to some sort of absolute standard I don't think that one exists. As several commenters note, the grading is where the real work lies. Are you looking for mass-market explanations or discoveries of off-the-beaten-path caverns no one knew existed? I suggest that of greater value than providing a Procrustean bed of unified definitions would be a clearing house of information - in mass-market language perhaps - describing for prospective students, journalists, and employers which nooks and crannies of economics each of your exam candidates likes to explore, tying it back to an overall "map" of the field.

Having them answering:

Do you agree with the statistics and theories who have regulators, with their risk weighted capital requirements, opining that what is ex ante perceived as risky, is ex post more dangerous to the bank system than what is perceived as safe?

Could perhaps help to rid us from the mother of all regulatory mistakes.

What did risk-weighted capital requirements do to you?

Nothing on Earth is ideal. I agree that the Basel I, II, and III capital "programs" are not optimum. I preferred plain-vanilla leverage ratios, adjusted for unrecognized loan losses, analyzed for levels and trends.

I think it's "myopic" to call R-WC "the mother of all regulatory mistakes." In addition to capital adequacy, regulators seriously look at levels and trends in asset quality, management (most important), earnings (reflects management effectiveness/efficiency and solvency - first line of defense for asset losses), liquidity, and sensitivity to market risk.

My experience is (for years they postponed implementation) that the new "current expected credit loss" (CECL) methodology for loan losses is causing more angst among bank managements.

Can you confine it to three, not 17, reasons?

Make it happen, Professor Cowen. Society would benefit if academia made more of an effort to communicate with the masses, to share their expertise and (hopefully) passion for their subjects. The only problem is the masses won't read 20 pages. In fact, the masses don't care and that's a big part of the problem. Oh well, maybe a consolation prize could be the removal of departmental walls and biases in academia.

Let’s modify the grading scheme. First remove the length requirement. Then give the essays to a front line supervisor in a company. Ask them to put the essays in three piles. The first pile is for those they do not understand. Those get an F. The second pile is for those that they understand. Those get a C. The third pile is for those that they would actually use to help them make a better decision in their business. Those get an A.

The problem with this "exam" is that every piece of writing is directed to an audience. Is the essay targeting experts or students?
Should it offer a novel perspective or summarise the existing perspectives? Without defining an audience, this exam is meaningless. Still fun to think about, though.

I think that would be a very difficult exercise. Many of what may be the "right" answers to those questions involve some logical reasoning, much of which can not be supported by empirical evidence. Thus, a dollop of ideology may be a required part of the answer, and that then can get students into trouble with their professors. For example, my answers to 2 of your questions would look like this:

Why did China do so well: Mainland China has a history of cultural norms, education and institutions that are effective ingredients for economic growth. These ingredients were counteracted by a series of negative events, including war with the Japanese (circa 1930), civil war (circa 1940s) and government oppression (1950s - circa 1975) that seriously retarded the economic growth, but which did not damage the underlying culture, educational system and institutions. Thus China was able to experience rapid economic growth post circa 1980 once these negative factors were eliminated (in the case of war) or ameliorated (in the case of government oppression). Post 1980 China was thus able to "bounce back" towards economic output per capita that is roughly similar to many of its neighboring countries.

Even worse is your question about healthcare, which having been in the field for 2 decades and thus being very opinionated on this subject I would answer: "It is very difficult to derive a system in which hospitals compete substantially more effective on a fundamental basis given their underlying dependence on the third-party payor system. Under the third-party payor system, the persons who receive care are, for the most part, distinct from the parties that pay for the care. Because of the gross misalignment of incentives, the underlying care in the United States is of reasonable quality, but the cost is substantially higher than it otherwise would be. Specific hospitals can significantly improve the quality of their overall value proposition (i.e., a combination of prices and quality of care) by specifically catering to "private pay" patients. But only hospitals with specialties that do not rely upon high third-party payments (e.g., cosmetic surgery) and hospitals that otherwise have a higher than typical blend of private pay patients are in a position to implement such strategies. Fundamentally, hospitals and the entire U.S. healthcare system would likely benefit from the elimination or at least a reduction in the third-party payor system. This in turn would likely require a modification to the U.S. Federal income tax law such that employer payments for health insurance and payments by individual persons for health insurance and healthcare expenses are treated equally for purposes of deduction from taxable income." Many professors would strongly disagree with the basic premise of my answer, and thus they would flunk me!

Here are my exam questions:

1. How many illegal migrants are raped trying to enter the US in a typical year? (Hint: About 1 million migrants will try to jump the US southwest border this year. The Border Patrol will stop at 40,000 per month, of which 25% will be women.) True of false: This is the direct result of US immigration policy. What kind of problem in economics would we say is this kind of phenomenon? Does economics have an unambiguous prescription for the problem? What percentage improvement would we expect to see in related rape statistics if your solution were implemented?

2. Some argue that the declining marginal utility of wealth and income suggests that the way econ grad students are taught to write and structure papers is all wrong. Why might this be? How does it relate the public's perceptions of sermons at church over the last half century?

3. Adam Smith argued that men are motivated by self-interest. Are politicians motivated by self-interest? If so, how do we characterize that self-interest? If not, why not? How does corruption factor into the equation? Does 'public service' really exist, or it is merely a balance of votes?

and who would be able to judge?

Great idea, but why not make it 10 pages instead?

Great post. "How much should our profession be focused on being able to write good answers to such questions?" Is this a call for economics professors to be able to interact with their core ideas or is it a call for them to be able to write? It seems like the former, but I ask b/c in "Ideas Industry," Drezner notes that is considered an especially talented writer amongst scholars. (I'd describe him as competent.) I also note that in "Wild Life," Trivers rates Drury as one of his greatest influences, but he also notes that Drury was underrated in his field (biology) because he wasn't a strong writer. Finally, I also take this post as wondering over the extent to which professors' quality is determined by their ability to lecture, particularly to a 100 level course, w/o losing the audience. It seems possible to me that the best answer is some form of "it depends."

A few other commenters have noted that economics, maths, and sciences professors are generally not very strong writers who would benefit from a reading of Strunk & White. Surely the worst writers, however, are postmodern English professors. What's the ratio of useless words to useful -- 5 to 1?

Over all the academic fields, it's of course going to be the philosophers who will be the best at this, since this is the very task they need to do to even get into grad school and it's the skill they spend honing over most of their career.

Generalizing from that, in economics, I guess folks who are closest to philosophy would do the best, such as people working in normative economics, law and economics, economic justice, economic methodology, decision theory, and so on.

Of course people far away from philosophy, such econometricians, mathematical economists, experimental economists, monetary economists need not necessarily do poorly, since they could've obtained their writing skills by other means. But you might expect them on average to do worse than the first cohort.

Why did China do so well? At what? Making junk?

Yes, making Chinese junks.

I am typing this answer on a beautifully made Apple computer that was manufactured in China. Any questions?

great post ... (inside blog trivia observation follows) every once in a while there will be a commenter on a blog with a decent comments section (this one, for example, most days) writing what is, for her or him, an incommensurately real long comment - not 20 pages long, but a 2 page comment (this comment is almost 3/4 of a page!) looks huge in a comment section like this, like those Beluga whales that used to swim around that little tank at the Brooklyn Aquarium (as Casey Stengel used to say, you could look it up). Back in 2011, I wrote the definitive explanation - on a blog that is now only accessible through the Wayback machine - of what George Steiner was trying to say when he expressed regret for not having read Sarpi on the Council of Trent. Back in 2004, I wrote a long comment on why - out of the Cowsills, the Brady Bunch, the Partridge Family, and the Jackson 5, only the Brady Bunch described a real family in a real world. Just 2 pages - not 20 - but probably nobody read it on the small blog I commented on (the blog's main theme was not poetry: had it been, I might have invoked Wallace ('my friends call me Chief') Stevens - "Like an evening evoking the spectrum of violet ...A woman writing a note and tearing it up ...the palm stands on the edge of space [remember where they filmed those shows, and the Southern California movie, not as good as the shows but the cinematography, standing alone, was exponentially better] - the wind moves slowly in the branches ... the vernacular of light, omitting reefs of cloud, empurpled garden grass ....[beach houses that I could not afford in summer and that were not for rent in winter]

It will be about a 15 minute task for the AIs of even a non-AI dominated future to roll out about a thousand of those 2 page answers, with full annotations.

His friends did not call him Chief, I made that up. He never visited Ojai or Livermore or Pleasanton and maybe never visited Thousand Oaks, but he wrote this:

"like an evening evoking the spectrum of violet ...A woman writing a note an tearing it up .... the palm [tree] stands on the edge of space ... the wind moves slowly in the branches .... the vernacular of light, omitting reefs of cloud, empurpled garden grass ...."

He visited Paris but not Lyon and there are thousands of people with paintings that he enjoyed and may even have talked to the reviewers about (he was a huge fan of the fashionable modern paintings of 1910- 1935) hanging on a wall in their nice houses they don't know it I doubt the specialist in that era of modern art at Sotheby's knows it most of the time, if you don't have a nice house you can see a few of them on the walls in the nice apartments in old Doris Day movies, even in a Loretta Young movie or two.

Chicago would win of course :)

This is a pretty interesting question. You should add a poll and we'll see who would win.

This essay idea is a good one for law professors too!

Comments for this post are closed