Skip to content

Author: Andrew Samwick

Published on

I occasionally receive inquiries from high school students and others who are just beginning their study of economics and want to know how to get started on economic research. My standard advice has four recommendations:

  1. Read a good introduction to economic reasoning for non-specialists. The best one I have found is my colleague Charlie Wheelan’s book, Naked Economics.
  2. Build a foundation in three areas -- principles of economics, calculus, and probability/statistics. For those without access to the relevant AP courses, Khan Academy is a good substitute. Take a look at microeconomics, macroeconomics, calculus, and statistics.
  3. Acquire some proficiency in computer programming. It seems that most students are learning the basics in Python. There are plenty of online resources to pick that up. (And for advanced students, see QuantEcon by Thomas Sargent and John Stachurski.) Once you have the basics, you will want to switch to a language that is meant for data handling and statistical analysis. I learned on Stata (Version 0.92!) and still use it, but many people are now learning on R, which is freely available. Here's the place to get started.
  4. In some respects, those are the easy recommendations. They build your analytical and computational toolkit. That will answer some of the “what” and the “how” of economic research. But the most important question is the “why.” Good research is based on an interesting question. What questions interest you? What assertion of fact or causality have you heard in public discussions about which you are skeptical or for which you require evidence? That’s the genesis of a good research project. So read widely about the economy and society. As I discuss in this inaugural episode of Big Green Economics, economics gets better when mixed with other disciplines.

Published on Leave a comment on Looking Ahead to 2024

David Brooks pins our current social and political problems on Collapsing Levels of Trust in his recent column in The Atlantic. I think he gets most of his arguments right. Without evidence that people are in general trustworthy, individuals will tend to congregate with others in their tribes. The way this is playing out today is polarization toward the far left and the far right.

As Rabbi Jonathan Sacks puts it in his recent book, Morality: Restoring the Common Good in Divided Times:

The far right dreams of a Golden Age that never was, and the far left dreams of a utopia that will never be.

It is not an unreasonable reading of 20th century history to say that when the extremes are emboldened in pursuit of these dreams, the result is violent and horrific. What are reasonable people supposed to do?

In brief, form a new tribe in the center. I am politically to the right of the center, but probably no more so than most of the people whom I call friends are to the left of center. If we chose to, we could waste a lot of time on Facebook, Twitter, or other anti-social media arguing about whether the lunatics well to their left are worse than the lunatics well to my right. We would be making the problem worse, by acting like partisans and letting the extremists set the agenda for public discussion and civic life. And yet this is the pattern into which we have fallen. Instead, we could disavow them both. 

 

I would wager that a coalition that included them on its left flank and me on the right flank would constitute a majority and would enact sensible policies. Centrist parties are thought not to be possible in our Presidential system in part because the parties that start on the far left or far right will find it easy to encroach on the center without losing the support of their flanks. I understand this as an equilibrium result. Are we in or moving toward such an equilibrium? I see no evidence that the two major political parties are courting the political center by moving toward the center. So perhaps there is a window of opportunity during which this coalition could govern. In other countries, they would call this a unity government, and such governments typically get formed when the alternatives are pretty awful. I am surprised that 2019 was not awful enough to spark such an insurgency of the rational. I suspect 2020 is that awful but the timing relative the Presidential election is unfavorable.

 
The last link and the phrase it highlights are to a TEDx talk by my colleague Charlie Wheelan, who put forth a sensible platform in the Centrist Manifesto in 2013 and founded Unite America to try to get a pivotal centrist bloc in the Senate. The Achilles heel in that strategy turned out to be that it presumes a Congress that seeks to perform even its most basic Constitutional responsibilities. So, looking ahead to 2024, we should be thinking about a centrist ticket for the Presidency, composed initially of a bipartisan duo with a prudent and responsible platform that begins to address the many crises we have built for ourselves. If the ticket wins, then the party can immediately recruit Senators and Representatives from the political center for a majority in each House, leaving the two existing parties in the extremes. Addressing those crises, even if incrementally, is what will help rebuild trust in our public life.

Published on Leave a comment on 9/11 for Our New Arrivals

This fall marks the arrival of the Class of 2024 on college campuses. This is the first cohort in which most members were born after the attacks of 9/11. A strange milestone. To honor the memory of those who died, and to help inform young adults of the events of that day, Keith Hennessey has crafted a study guide, available here. Keith was working on the Hill at that time and acknowledges that the passengers of Flight 93 may have saved his life. I had the good fortune of meeting Keith a few years later when my time at the Council of Economic Advisers overlapped with his time at the National Economic Council.

Published on Leave a comment on DC Statehood

Last week, the House passed a bill to allow for statehood for Washington, DC. I am in favor of representation in national legislature for the residents of the District, but I am not in favor of statehood. The Founders intended for the District to not be a part of any state. Here is a proposal formed from ideas I've picked up over the years from various sources, so I don't claim originality here.

We could think of the District as part of Maryland, which gave the land for it. Allow the District to be its own Congressional district. This is only fair, as the House allocates representation based on population. The District gets a minimum of 1 Representative (and is protected from gerrymandering that might occur if it were formally part of Maryland). For the Senate, allow residents of the District to vote for the Senators that represent Maryland. Then its residents have two Senators to represent them. 

This solves the problem that there are no voting members of the national legislature who are responsive to the residents of the District. It does so with minimal disruption. I acknowledge that Maryland has to take one for the team and needs to consent. But this is where I would start the conversation.

Published on Leave a comment on Corona Blogging

There are debates about what the policy response should be to the COVID-19 threat to the United States. I think there are three important steps that need to be taken, beyond what the Federal Reserve is already doing to support the banking system.

One, rather than spraying $1000 checks around like confetti, the Federal government needs to re-insure and augment state unemployment insurance programs. Yesterday's report showed that we had a jump in initial UI claims, from 211,000 in the week ending March 7 to 281,000 in the week ending March 14. Expect that to go much higher for this week (released next Thursday) and next. In my view, Federal support for state unemployment insurance programs is the best way to inject cash into the economy. Top off replacement rates at higher levels, since these claims are due to factors beyond the control of workers and their employers, and extend them for longer than the current claiming period. Make the funds available to states in proportion to their working-age population, to avoid explicitly bailing out states that were not properly funding their systems. What's important here is the Federal government's announcement. It is the announcement that sets people's expectations about their resources and these expectations that drive consumer behavior.

Two, we need to recognize the manpower cost of successfully addressing the COVID-19 threat and mobilize accordingly. I encountered two useful accounts that shaped my thinking about this. The first was a Ted Talk by Bill Gates from 2015 that resurfaced recently. We should have taken his advice then (in the wake of the Ebola outbreak). We should start the mobilization now. If the government is going to spend, it should use this occasion to buy what it needs. In bulk. (Sound familiar?) The second was a first-person account from an economics colleague who flew from London to Beijing, posted to Twitter. What struck me about his account was how labor-intensive the testing and quarantining process is in the places where that process is being applied rigorously enough to actually contain the virus. And now we have a whole bunch of service-sector employees looking for temporary work. State and particularly local governments should be hiring and training, with funding provided by the Federal government, legions of people to facilitate this process. When the immediate threat has been contained, that corps of people can be shifted to address long-term preparedness along the lines that Gates was describing.

Three, for any publicly traded company that needs short-term funding to remain liquid and solvent, like airlines and hotels, the Federal government can provide it. However, it needs to take warrants as part of the deal, similar to the arrangements for banks in TARP, so that it more fully shares in the eventual recovery (and signals its belief that there will be a recovery). This is not an appropriate time to advance clearly partisan objectives when the business disruption is due to an external factor like the virus.

Published on Leave a comment on The Management Consulting Business

Christopher Ingraham, writing in The Washington Post, links to an old working paper of mine in his article, How consulting companies like McKinsey optimized American inequality. The paper began as my co-author, Ajay Prakash's senior thesis, which I advised and then helped revise. This topic is in the news of late due to Pete Buttigieg's experience working at McKinsey and the perceived need by the chattering class to demonize just about everything on everyone's resume. I currently have no intention of casting a vote for Mayor Pete in 2020, but I don't hold his McKinsey experience against him.

The subtitle of the article is, "Two things tend to happen when businesses hire a management consultant: Stock performance rises and payroll falls." To be clear, the announcement effect is negative but the long-term effect is positive risk-adjusted returns on a scale that more than offsets the bad news at the announcement. That shows that whatever happens after the consultants are brought on board, it tends to turn the company around. The article should suspend its focus on distributional issues between shareholders and workers long enough to acknowledge that this is consistent with the consultants adding value to the economy.

What I found frustrating about the article is that it doesn't seriously engage with the counterfactual. Our paper also found that "announcing companies earn negative risk adjusted returns prior to their announcements." This means that the company was underperforming, and that underperformance was going to have be addressed, whether through the consultant's advice and actions or some other way. Imagine the firm failed instead. The workers who lost their jobs would be in the same position, but every other worker is better off due to the consultants. To give some sense of the scale of the two groups, we found that the median employment growth (relative to the industry) was -5% for firms that announce they've hired a management consultant. Is it really unacceptable to lower employment by 5% if it means shoring up the other 95%?

I would say 'no.' What I would say as well is that I would like to know that the workers who are displaced are treated fairly, whether through unemployment insurance or some other mechanism. For example, publicly traded companies that lay off workers under such circumstances might be required to offer severance in the form of warrants (stock options issued by the company to those workers). If the company does well as a result of the layoffs, then those who were laid off share in that improvement.

Published on Leave a comment on Sorting Out Discipline in the NFL

Like everyone else, I was shocked at the video of the altercation at the end of the Browns-Steelers game on Thursday night. As bad as it was, it could have been much worse. When Myles Garrett swung Mason Rudolph's helmet at the latter's unprotected head, he didn't hit him squarely or with the hardest part of the helmet. So Rudolph has no sign of a head injury that could have been catastrophic. (This by itself should eliminate discussion of the punishment extending into criminal or civil lawsuits.) The melee also didn't extend beyond the players on the field. Kudos to stadium security for keeping order in what must have been a very challenging environment. Garrett has been suspended indefinitely, which is appropriate, as are the lesser suspensions for the Steelers players.

There are two aspects of this unfortunate turn of events that have not received much discussion, and they both pertain to the coaching in this game. These remarks are not intended to lessen the blame on the players involved but to point out that player behavior is shaped by coaching decisions and the environment coaches create and the culture they promote for a team. If the NFL, and the ecosystem of commentators that follow it, want to send a message that this sort of player behavior is unacceptable, then they need to assign blame and discipline as broadly as is appropriate.

First, Mike Tomlin and the rest of the Steelers' offensive coaching staff should be embarrassed at the play call in question. Look at the video. Garrett is unblocked on the play, to free up the lineman to block for the running back who catches the short pass in the backfield. There are under 20 seconds left in the game, victory is beyond reach (and brought no closer by this play call), and your team has had a rough night. Hand off the ball to the running back, let the game clock expire, and get ready for the next one. What on Earth were they thinking to leave their quarterback unprotected like that? Even if the play is successful in terms of yardage, the quarterback is going to take a punishing hit. That's a stupid tradeoff to make, and one that shows little regard for Rudolph's safety.

Second, others have commented that the Browns are a bit of a train wreck this season. This altercation is just the latest and worst manifestation. What is the NFL to do? They have to assign blame, and through that assignment, discourage future behavior of this sort. That means holding coaches accountable for their players' on-field behavior. Were it up to me, I would also suspend Browns head coach Freddie Kitchens, perhaps along with his defensive coordinator and defensive line coach, for several games. The coaches set the tone. They determine which behavior is rewarded and which is not. They failed to do that properly here. Going forward, a policy of suspending coaches for their players' behavior could have the effect of changing the culture and environment, or even personnel decisions, in the interest of avoiding the most egregious behavior on the field.

Published on Leave a comment on In Praise of Stan Collender

I learned today that my friend and former co-blogger Stan Collender passed away on Friday. James Hagerty has a thoughtful obituary in The Wall Street Journal today.

It was very early on in my blogging that I discovered Stan's writing on the budget. There are many writers who are deficit hawks at heart and know about the budget in fairly general terms. Some of us can even write coherently for an audience on budget matters. Stan was a fellow traveler, but he brought so much more to the discussion -- an impressive command of the details, a keen sense of the political strategy underlying "Budget Battles," and a wit as sharp and quick as any you will find. He will be missed.

Published on Leave a comment on Fiscal Recklessness, 2018 Edition

I was invited to present a lecture at Middlebury this past week on "Income Tax Policy and Economic Growth." I wanted to present a picture of just how bad the projected federal deficit is for the next decade. The following graph, which shows the last 50 years of data on the unemployment rate and the federal deficit, is what emerged:

Sources: CBO, BLS

The horizontal axis is the unemployment rate, as a percentage of the labor force, and the vertical axis is the federal deficit, as a percentage of GDP. Other things equal, we would expect the line to be downward sloping -- the higher is the unemployment rate, the worse will be the deficit, as both automatic stabilizers and additional stimulus efforts kick in. Each blue dot represents the year indicated over the last 50 years, 1968 - 2018. The best-fit linear relationship indicated by the blue dots is about -1.2: for every additional percentage point of the labor force that is unemployed, the federal deficit widens by about 1.2 percentage points of GDP.

Departures from that general tendency indicate budget policy that was unusually strict (toward the Northeast direction) or unusually loose (toward the Southwest direction). The orange dots are the next 10 years of projections, based on CBO data. Note that there is no recession projected in these data -- the unemployment rate is projected to remain below 5 percent over this decade. Despite this rosy economic projection, the deficit is always about 4 percent of GDP or larger. We will have never had deficits that large as a share of GDP with unemployment that low. Relative to where we were 20 years ago with similar unemployment rates, we are running deficits 5-6 percentage points of GDP larger. That's President Trump's fiscal record.

That is the mess we are in. My main concern is intergenerational equity -- there is no ethical reason to pass along the burden of financing structural deficits to future generations. And then, eventually, there will be further concern if and when the rest of the world decides it is no longer content to allow us to roll over our debt at such favorable interest rates.

Published on Leave a comment on Bad Journalism, Bloomberg Style

Some days, you just have to wonder. What could be the motivation for an article in Bloomberg news that starts out like this?

U.S. Students Spend More Time Working Paid Jobs Than Going to Class

Facing mounting debt, U.S. college students spend double the time working paid jobs than in the library. 

The rest of the article goes on to explain how awful it is for students, what with all the work and borrowing. But the proof of this assertion is a survey by HSBC in which students were prompted with the question, "On average, how long do you spend doing the following each day?" Here are the answers:

Going to lectures/tutorials/seminars: 2.3 hours
Visiting the library: 1.5 hours
Studying at home: 2.8 hours

That's a total of 6.6 hours on school work.

Working (paid employment): 4.2 hours
Volunteering (unpaid): 0.9 hours

That's a total of 5.1 hours on work unrelated to school. So yes, based on these categories, 4.2 > 1.5 and 4.2 > 2.3, so the facts asserted in the headline and the statement that follows it are true. But 5.1 < 6.6, so what's the big deal?

But wait, there's more. The same survey reports the following responses for the rest of the day:

Texting/messaging/emailing: 2.3 hours
Watching streaming devices: 2.2 hours
On social media: 2.5 hours

So that's 7 hours on screens, more than both school work and paid work. (Also listed are 4 hours of socializing.) I guess if I were going to write a news article in Bloomberg and be objective about it, I would include that. But then how could I make the case that life is somehow unfair to students because, what, between school work and other work, they only get to spend 11 hours socializing or looking at screens?

When I blogged about the media earlier this month, noting that with the increase in quantity we have seen some bad examples of quality, this is the sort of media I had in mind. A survey has been cited selectively and incorrectly to push a point of view. Further, I first came across this in my local paper, which had no link to the survey. This is bad journalism. It is unprofessional. And it erodes the trust we'd like to have in those who bring us the news.