Jason Furman's Reviews > Good Economics for Hard Times: Better Answers to Our Biggest Problems

Good Economics for Hard Times by Abhijit V. Banerjee
Rate this book
Clear rating

by
4651295
's review

really liked it
bookshelves: nonfiction, social_science, economics

This book has a huge amount of good economics. It surveys a wide range of areas: labor, tax, growth, politics, immigration, trade, and generally provides up-to-date discussions of some of the latest literature. The discussions of development--particularly India--are subtle, nuanced and thought provoking. A lot of the evidence is in the form of randomized control trials (RCTs), Abhijit Baanerjee and Esther Duflo are as committed to the method and process as they are to any particular conclusions. They are also consistently skeptical of neoclassical theorizing on the basis of first principles, and particularly critical of the ideas of people's preferences as having special weight (a la Becker and Stigler) because they see these preferences as shaped by society and malleable.

Banerjee and Duflo do not limit themselves to RCTs and attempt to discern what can be learned without them on a variety of topics, although they mostly throw their hands up at the impossibility of understanding some of the big questions in growth--like what other countries would need to do in order to achieve Chinese-levels of growth.

The book also has an interesting shifting back and forth between the advanced economies and developing ones, drawing out similarities but also contrasts between them that help better understand each. Often this technique and presentation works, but sometimes it does not and creates a slightly disjointed feel.

What was disappointing to me--and I say this relative to my high expectations based on the authors, their research, and also their previous book Poor Economics: A Radical Rethinking of the Way to Fight Global PovertyPoor Economics--was that when it came to the biggest contemporary hot button issues they left all nuance and care behind to present selective evidence for an overly simplistic narrative that would be comforting to typical progressive reader without challenging any of their preconceptions. Moreover, they framed their view too much in terms of "good economics vs. bad economics" and the "facts" when it was actually a very substantial dose of values that were in play (many of which I agree with) along with a subset of the facts that would be more suitable to a polemic than a self-styled balanced presentation.

Four examples:

1. On immigration, they present a 100 percent positive picture while denying that there has been a very rapid and in many countries unprecedented change in the composition of the population. I happen to think this change is wonderful, but understand that many others do not--and their perception of the facts is not entirely wrong. Banerjee and Duflo, however, essentially deny these facts writing, "MIGRATION IS BIG NEWS, big enough to drive the politics of much of Europe and the United States. Between President Donald Trump’s imaginary but enormously consequential hordes of murderous Mexican migrants and the anti-foreigner rhetoric of the Alternative for Germany, the French Rassemblement National, and the Brexit crew, not to mention the ruling parties in Italy, Hungary, and Slovakia, it may be the single most influential political issue in the world’s richest countries… Why the panic? The fraction of international migrants in the world population in 2017 was roughly what it was in 1960 or in 1990: 3 percent…. Racist alarmism, driven by a fear of the intermingling of races and the myth of purity, doesn’t heed facts."

The "facts," however, from the same source they cite for the 3 percent statistic is that the foreign-born share of high-income countries has risen from 7.6% in 1990 to 14.1% in 2017 with even faster rises in several of the countries they cite. "Facts" and "economics" do not settle the debate about whether this change is wonderful or threatening and demeaning the people who find it threatening as factually delusional does not help build understanding, change minds, or advance good economics.

2. The discussions of inequality have a similar feature. Words like "exploding inequality" and "hard times" overstate what the data say. A wide range of data show that a inequality is high and has increased, but Baanerjee and Duflo rely too uncritically on just one source and perspective on the data (Piketty, Saez and Zucman), missing out on the nuance they bring to other areas of discussion. In the process, they don't address the methodological choices made by these authors or present any facts that run counter to their narrative (e.g., CEO pay relative to median worker pay has been flat for two decades now, the very richest Americans are less likely to be inherited wealth than in the past, mobility is higher at the bottom of the income distribution than the top, income of the bottom quintile or the median is very highly correlated with average incomes across countries, the fiscal system has become more redistributive, etc.)

Overall, the term "dark times" that a is the subtitle and leitmotif of their book seems somewhat off for the United States (when in the past would rather be alive than today?) and is particularly off for a book that devotes substantial fraction of its focus to India and other developing countries, places that are growing rapidly and have lifted massive numbers of people out of deep poverty (points they also make in the book).

I have no dispute that we could and should do a lot better in the United States and around the world (inequality is too high, income growth and mobility too low, etc.), just wished the evidence was a presented with more nuance.

3. The discussion of taxes is a useful corrective to the wild-eyed claims of many conservatives about the huge growth potential of cutting taxes. But in the course of this corrective, the authors' almost appear to deny that incentives matter at all, that taxes can affect welfare and macroeconomic outcomes, and the like. Some of the evidence they present for these propositions is high quality, as in top journals. But others were observations like tax rates were high in the 1950s and growth was high, ergo taxes don't hurt growth, statements that do not meet the evidentiary standards they set elsewhere and that express more confidence in a precise estimate of zero on the impact of taxes as opposed to failing to reject. a null of zero or even the magnitude of uncertainty they appropriately express in the chapter that discusses growth.

4. My fourth example is much less clear cut, more of a disagreement than a clear cut issue. But I think Banerjee and Duflo overly caricature trade economics as an apologia for trade that has almost entirely ignored inequality, something I think is too simplistic a presentation of a field that has its modern theoretical origins in Heckscher-Ohlin and Stolper-Samuelson which are all about inequality (and discussed extensively in Good Economics). Moreover, I think the authors' place too much weight on the Autor, Dorn and Hansen China Shock literature which I view as largely about the gross costs of trade not the net costs of trade because they fail to capture the general equilibrium effects and the tremendous gains from exports. Moreover, it is not clear if the China Shock tells us about China's entry into the WTO or China's growth, likely it is more the later than the former--which matters because it is more about developments in China than policy changes in the West. I also think their discussion of the "dirty little secret" that the gains from further trade liberalization are small in advanced economies puts too little weight on the fact that standard static Ricardian models may miss gains that come from increased competition, a larger market for innovation, learning by exporting, increased specialization in innovation, supply chains, and regulatory reciprocity or even harmonization. I don't know how big the gains from all of these are, they do not lend themselves to precise modeling, but it is possible they are substantially larger than the Ricardian effects that do lend themselves to precise quantification.

The above really emphasizes everything I disagreed with, but I want to reiterate what I began with--much of the book is a fabulous, sophisticated, exciting discussion of economics and what it can teach us about some of the most important issues in the world today. These criticisms are out of respect for the mission the authors accomplish in much of the book--to present a balanced picture of the facts, the good economics that interprets these facts, while also being honest about our uncertainty and the limits on our understanding.
54 likes · flag

Sign into Goodreads to see if any of your friends have read Good Economics for Hard Times.
Sign In »

Reading Progress

November 18, 2019 – Started Reading
November 23, 2019 – Shelved
November 30, 2019 – Finished Reading

Comments Showing 1-4 of 4 (4 new)

dateDown arrow    newest »

Scott Tennican Superbly written review.


message 2: by Samuel (new)

Samuel Gray Jason Furman is one of my favorite economic policy thinkers.


message 3: by Doris (new) - added it

Doris I must come back and re-read this review after I have finished the book. Thanks for bringing up alternative things to consider about this subject!


Mohib Half way through the book and I agree on most of the points you make.


back to top