Dave Donaldson has won the John Bates Clark Award

by on April 14, 2017 at 3:35 pm in Economics, History | Permalink

Here is the award citation, here is one excerpt from it:

Donaldson’s paper “Railroads of the Raj: Estimating the Impact of Transportation Infrastructure?” (American Economic Review, forthcoming) investigates the economic benefits from building transportation infrastructure studying the case of railways in 19th century India. This paper is widely viewed as both a methodological breakthrough and substantively important paper in the field. Donaldson assembled a new and rich data set from archival sources about the expansion of railroads in India through the 19th and early twentieth century and the volume of inter-regional trade in the same period. He then uses the data to look the effect of access to railroads on real agricultural incomes. To check that this effect does not come from building railroads where the growth was predicted to be, he uses the fact that a number of proposed lines did not get built or did not get built when they were proposed to be built. Assuming that the proposal was based on what the contemporary experts thought were the areas of greatest demand for transportation, these un-built railroads should also have an effect if they were any good at predicting growth. He finds no such effect.

The second part of the paper builds a quantitative model where the effect of trade on real agricultural GDP is fully captured by one sufficient statistic: the share of expenditure that each Indian district allocates to goods produced in the district. When that share is low, it indicates that the relative price of imports in the district is low, and in turn, that the welfare gains from trade are large. Controlling for shocks to technology (mainly rainfall in this case), he finds that observed changes in real GDP following access to the railroad move almost one for one with the sufficient statistic predicted by the model, thereby making the case that the benefits of the railways is indeed the result of increased trade.

There is much more of interest at the link.  Here are copies of the papers, overall I am delighted to see a Clark Award that so prominently features economic history, not to mention India and trade.  Donaldson is at Stanford, here is his home page.  An excellent pick, but this one was a surprise to me.

1 Scott Mauldin April 14, 2017 at 3:46 pm

“Assuming that the proposal was based on what the contemporary experts thought were the areas of greatest demand for transportation, these un-built railroads should also have an effect if they were any good at predicting growth.”

This seems spurious to me. He may expound in the book. but on the surface there could be hidden causal factors both retarding economic growth and preventing railroad construction in a given area, for example changing patterns of trade or population movement.

2 Steve Sailer April 14, 2017 at 4:01 pm

You could do a similar analysis of which planned freeways got built in Los Angeles and which didn’t. For example, the Mulholland Freeway along the crest of the Hollywood Hills was planned but never built, for a bunch of reasons, one of which was diminishing marginal returns: the parallel east-west 10 (Santa Monica) and 101 (Ventura) freeways were built so there was less need for a third freeway running in the same direction in between them.

3 Nahim April 18, 2017 at 11:46 pm

He does expound in the paper, which I just had a look at. It seems proposed railroad projects could be cancelled at several different stages. If railroads were located based on expected economic growth, then you’d expect projects cancelled at a later stage to be in areas that experience higher economic growth. For example, let’s say a railroad is never proposed in region A; it gets proposed but shot down in stage 1 for region B; it gets proposed but shot down in stage 4 for region C; and a railroad gets built in region D. If the effects described in the paper are driven purely by how these regions were selected, you’d expect growth in D>C>B>A. Instead, the paper finds growth in D>C=B=A (consistent with the railroads having an impact): there does not appear to be a systematic difference in the growth rates of A, B, or C.

Donaldson discusses how railroad plans were passed. Very often there were competing plans proposed by engineers/groups, each of which would affect a different set of areas. Reading the historical account, it would seem there was a significant random element as to which plan actually passed through (for example, one plan was cancelled once the Viceroy finished his mandatory five-year term). That makes Donaldson’s strategy somewhat more believable.

The paper also provides some other pieces of supporting evidence. For example, some regions got a railroad because they were exposed to famine in the 1870s; by the time they received the railroad (in the mid 1880s), agricultural income had largely rebounded. There were other regions exposed to famine that did not get a railroad post-famine because the idea of building railroads as a way to combat famine was popular only at a particular stage in India’s history.

Of course none of this conclusively rules out a spurious effect. But my impression is that the author did a careful job of thinking about the various ways in which the estimates might be spurious, and providing evidence to rule many of those ways out.

4 Steve Sailer April 14, 2017 at 4:02 pm

“thereby making the case that the benefits of the railways is indeed the result of increased trade.”

Did anyone expect something else from building railroads? Chuck Berry backbeats?

5 The Other Jim April 14, 2017 at 6:17 pm

Well, it’s easy to get confused these days about railroads actually bring to people.

Today, Dems promise that “high-speed rail” will bring about the death of the obviously useless automobile, while providing virtually unlimited high-paying union jobs. There is literally no downside.

“Increased trade”? Are you kidding? Did you think that was on the list?

6 Steven Sailer April 14, 2017 at 6:41 pm

I’ve always wanted to go from somewhere outside Bakersfield to somewhere outside Fresno at 200 mph and soon* I will be able to!

* “Soon” on a geologic time-scale.

7 Kris April 15, 2017 at 8:29 am

Why not? That stretch of road is quite dismal, unless you make a detour to Sequoia or King’s Canyon parks.

8 Donald Pretari April 15, 2017 at 11:01 am

You could visit Hanford, my hometown. Get a milkshake at Superior Dairy.

9 Elan April 15, 2017 at 12:21 am

A lot of left wing historians seem to think that infrastructure just lets the capitalists carry the spoils of empire away, especially in times of famine. It’s interesting to see rigorous work showing that isn’t true.

10 rayward April 14, 2017 at 4:38 pm

Of course, the Brits didn’t care about trade within India, only trade between Britain and India; ergo, Britain didn’t care about India transportation infrastructure, only transportation infrastructure between India and Britain. Is that relevant to America today? Does the hedge fund manager in NYC care about transportation infrastructure in America? Does the boy wonder in Silicon Valley care about transportation infrastructure in America?

11 Kris April 15, 2017 at 8:31 am

Isn’t the entire motivation behind the high-speed train from LA to SF that there is nothing (and nobody) worth anything that lies between?

12 leppa April 14, 2017 at 5:37 pm

…”This one was a surprise to me.”

Serendipity perhaps besides Alex’s posts from India , even TC has been posting about India and its economic history.

13 Ray Lopez April 14, 2017 at 5:58 pm

I don’t see what’s so special about this study. It’s well known that during the early days of the railroad, before diminishing returns kicked in, establishing a RR line helped farmers since it allowed them to export their crops to a larger market. It was true in the USA and no doubt India. It was also true with the Erie Canal (which apparently paid for itself in a couple of decades).

As for the brilliance of Donaldson, I prefer this game (IM Donaldson on the losing end, but still a brilliant game):
[Event “Kings Island,OH”]
[Date “1992”]
[White “Kaidanov, G.”]
[Black “Donaldson, J.”]
[Result “1-0”]
[ECO “D10”]

1. d4 d5 2. c4 c6 3. cxd5 cxd5 4. Bf4 Nc6 5. e3 Nf6 6. Nc3 e6 7. Bd3 Be7 8. h3 O-O 9. Rc1 b6 10. Nf3 Bb7 11. O-O a6 12. Qe2 Bd6 13. Ne5 Bxe5 14. dxe5 Nd7 15. Rfd1 Qb8 (15… Ndxe5 16. Bxh7+ Kxh7 17. Qh5+ Kg8 18. Bxe5 Nxe5 19. Qxe5 Rc8) 16. Nxd5 (16. Qh5 g6 17. Qh6 Ndxe5 18. b3 f6 19. a3 Qc7 20. Na4 Qd8) 16… exd5 17. e6 Nde5 18. Rxc6 Bxc6 19. e7 Re8 20. Bxh7+ Kxh7 21. Qh5+ Kg8 22. Bxe5 Qc8 ( 22… Qb7 23. Bxg7 Kxg7 24. Qg5+ Kh7 25. Rd4 Qxe7 26. Rh4#) 23. Bxg7 Kxg7 24. Qg5+ Kh8 25. Rd4 Rxe7 26. Rh4# 1-0

14 middle aged amateur veterinarian April 14, 2017 at 11:02 pm

Umm… is there not a whiff of the coffeehouse patzer at 16…? (not by both of course) And for the other Donaldson, he worked hard on many things, some of them special in their way. The world is a big place. Don’t rain on his parade, Ray! Everyone is wrong about a lot of things – I once talked with an IM type person who totally could not get the fundamental 2 dimensional simplicity of the Monte Hall problem or the even more basic simplicity of the slightly more modern problem [which I paraphrase as] ‘if there are infinite strings of numbers do they represent infinite realities that exist or infinite realities that include infinitely vanishing possibilities of other realities” and, I kid you not, believe it or not, the IM type person was convinced that Pascal’s wager was based on infinitesimals (for the record, it has Absolutely Nothing to do with small numbers versus large numbers) (in reality it is based on our awareness of two questions: why is there something instead of nothing or why is there nothing instead of something, ceteris paribus; and why are we so easily deceived into thinking that prime numbers are more interesting than other numbers (leaving out the undeceived among us: if there is a Ramanujan or even a Gauss reading this, or even a believably sub-Ramanujan or sub-Gauss rhetorician/undeceived know-it-all reading this, feel free to exult in your superiority to this comment).

15 middle aged amateur veterinarian April 14, 2017 at 11:07 pm

Kaidanov, for all his skills, probably could not paint “The Derby at Epsom”. (Gericault, not a chess player, to be sure, but he did not live very long, he might have taken up chess later in life) painted it.

16 middle aged veterinarian April 14, 2017 at 11:21 pm

The Gericault was an endgame type painting, check out Bonifacio Veronese (the other Veronese) for a great middlegame painting – Sacra Conversazione (a portrait of at least 8 people) – and I have not even started on the really good painters (I won’t start on them – this is just a comment thread, after all). (My favorite Sacra Conversazione is the one which is now in Vienna, by Lorenzo Lotto – check out the colors in the background and the details of consciousness in the portraits. I wouldn’t ever compare that painting to something as simple as chess. Maybe that is my ignorance.)

17 middle aged amateur veterinarian April 14, 2017 at 11:30 pm

And Ray, it was a special study. Read it. I tried to be a successful economist, back in the 80s, and could not figure out how to do it. Went into sales (of a sort), met a lot of nice people. But I don’t kid myself that the world of economics lost out when I left it.

18 middle aged amateur veterinarian April 14, 2017 at 11:31 pm

re 11:21 — (I won’t start on them) (except to mention the Lotto)

19 Ray Lopez April 15, 2017 at 11:43 am

“Umm… is there not a whiff of the coffeehouse patzer at 16…?” – what? 16. Nxd5!! is brilliant, so are you saying there’s a better defense than taking the knight, 16…exd5 ? (question mark to you, not to the move). Rest of your missive not read. If you can’t understand the chess move, no point in talking to you about your metaphysics my bean eating friend/fiend.

20 amateur veterinarian April 15, 2017 at 12:46 pm

Ray – the comment was satirical – it was, in rhetorical terms, a specific analogous echo to your first phrase (the “what’s so special” phrase) which showed a lack of understanding of the quality of the economist Donaldson’s “non-special” study. “Is there not” is not the same as “There is” and is often used to make the rhetorical point in question (I won’t bother with the Greek name – lo it up). Live and learn! I chose 16… as the highlight, didn’t think you would miss my intent.

21 amateur veterinarian April 15, 2017 at 12:52 pm

to make it clear the only part of the comment that was satirical was the coffee house patzer phrase. Lotto is a really good painter, and the Sacra Conversazione is not only Lotto at his best but it is the High Renaissance at its best. Look it up!

22 polyglot April 14, 2017 at 6:20 pm

Sadly, this proves nothing. There was substantial pressure on the District administration to make existing Railways ‘pay their way’. Indian statistics were highly manipulable. This is a worthless result because it deals with the opposite of a free=market case.

23 Ray Lopez April 14, 2017 at 6:51 pm

Ah, inspired by your post, here is what I think this paper is all about: it proves the opposite of the below, the opposite that these Indian scholars are claiming, that the British colonial railroad was looting. Now it all makes sense (otherwise the paper is redundant to what is already known about US railroads in the 19th century). Always remember economics is called “political economy” for something. Always follow the money, see which ax is being ground (except for our host TC, who is fair and balanced).

https://en.wikipedia.org/wiki/History_of_rail_transport_in_India (“While described by some as an example of the benefits of British colonialism, scholars such as Shashi Tharoor and Bipan Chandra describe the construction and operation of railways in India, initially by the British East India Company and subsequently by British companies hand in glove with the British colonial government, as one of the prime examples of British colonial loot. For example, companies were encouraged to overspend on such construction by the high rates of returns that had been guaranteed to British shareholders of these so-called Indian railway companies often based in London. “…each mile of Indian railway construction in the 1850s and 1860s cost an average of £18,000, as against the dollar equivalent of £2,000 at the same time in the US.” The extra costs were claimed from Indian people through added taxes and revenues.[4]”

24 Anon April 14, 2017 at 7:40 pm

Shashi Tharoor writes well ( I personally think his “Great Indian Novel” is a masterpiece) and came close to becoming UN Secy General , but for 1 Veto vote , but its hard to call him a scholar. His undergrad degree was in History but his PhD was in Law and Diplomacy , I think.

25 Ray Lopez April 14, 2017 at 8:23 pm

OK thanks for that. A quick review shows the GIN book is very colorful. https://en.wikipedia.org/wiki/The_Great_Indian_Novel (“Pandu is also enjoying his two sexually expert wives. While enjoying sexual congress with both at once…”, “The Seventh Book: The Son Also Rises” [prequel allusion to economist Gregory Clark’s book])

Bonus trivia: Greece and India both voted against the creation of Israel in the UN. Strange bedfellows.

26 polyglot April 14, 2017 at 9:48 pm

Greece and India, strange bedfellows? How friggin ignorant are you? Do you not get that the two nation theory of Islam vs what language you speak started with Greece? Turks who were Xtian got deported to Greece while Greek speaking Muslims went the other way.
India and Greece voted for the same friggin reason. They had had their heart torn out by Religion as constituting Nationality.
What is your major malfunction? Why troll people who know what they are talking about?

27 Anon April 14, 2017 at 10:28 pm

The Great Indian Novel is a take off on the Mahabharata.


The Mahābhārata is roughly ten times the length of the Iliad and the Odyssey combined,

Incidentally,Ray, how do you rate Kazantzakis ? He is in my top 5.

28 Kris April 15, 2017 at 8:41 am


Wasn’t it the Greeks who started that particular war against the Turks? And wasn’t it the Greeks who wanted to partition Turkish/Ottoman territory to gain maximum lebensraum? So it wasn’t an exact parallel with India, was it?

(Though I must say, I generally sympathize with the Greeks as being colonial victims of the Turks. But Venizelos and company made a bad miscalculation at the time.)

29 Ricardo April 15, 2017 at 1:12 pm

““The Seventh Book: The Son Also Rises” [prequel allusion to economist Gregory Clark’s book])”

Both Tharoor and Clark are paying homage to Ernest Hemingway’s novel “The Sun Also Rises.”

30 polyglot April 14, 2017 at 9:44 pm

Oh dear. Ray Lopez is a friggin idiot. I am a 54 year old Indian man who graduated in Mathematical Econ from the LSE at the age of 19. What is wrong with you Ray?
Why are you quoting non Economists like Tharoor? What is wrong with you?
Oh! Right. You are shite. Bipan Chandra is funny. But you are too stoopid to get why.
Have you ever made an intelligent comment? Nope. You are shite.

31 Sandia April 14, 2017 at 11:14 pm

Indeed we are seeing that the great accomplishments of macro are improved frameworks for backwards looking analysis. Not a bad thing!

32 Sandia April 14, 2017 at 11:18 pm

“Which way is the river going?”
“I can’t tell you”
“Why did it go the way it went?”
“That I can tell you, my friend.”

33 Barkley Rosser April 15, 2017 at 4:44 am

In the earlier India/colonialism thread I mentioned building RRs as one of the clearly good things the Brits did, only person to do so, I think. Anyway, I also recognize what has been mentioned here as well that many critics say that for longer term economic development at least some of the RRs should have been built elsewhere. But that they increase trade is certainly reasonable, even if in some cases the wrong sorts of trade. Those complaining the result is obvious should keep in mind these disputes and the fact that Donaldson has done some ingenious things with the data to try to get as clear an answer as possible.

I did not know of his work previously, and certainly did not forecast him, but he looks like a worthy recipient.

34 Kris April 15, 2017 at 8:47 am

I’ve been reading a lot of recent literature suggesting that freeway building through American cities in the 50s destroyed many urban communities. So building of infrastructure may not be a self-evident blessing.

Similarly, what I’ve read about railway building in 19th century India suggests that it made old local trade lines unprofitable, and impoverished settlements in the hinterland. This resulted in people moving close to railway lines and into bigger cities that were on those lines. So perhaps the railways did promote longer-distance trade within India, but this may not have been an un-mixed blessing.

35 Troll Me April 15, 2017 at 11:12 am

Maybe the unbuilt railroads were not built because someone else thought the economic rationale was not convincing?

Imagine if you give me a shortlist of 20 growth spots. I invest in 10. You observe the 10 that I did not invest in, and blame my original stupidity shortlisting them and not my wisdom in rejecting them after the shortlist.

Try again? He treats unbuilt railways as a random decision. Such decisions are not random.

36 Nahim April 18, 2017 at 11:55 pm

I think he does this a little better than what you are giving him credit for. Here is an analogy that’s closer to what he is doing in the actual paper (not the short summary reported above):

Imagine you give me a shortlist of 20 growth spots. I rule out 5 of them straight away because they are so unpromising. I do a bit of research into the remaining 15, and rule out 5 more spots. Then I do extensive research into the remaining 10, and choose 5 spots to invest in.

It turns out that the 5 spots I invested in perform the best. But strangely enough, the other 15 spots perform about the same. The 5 that I rejected out of hand do about as well as the 5 that I nearly invested in. That suggests that, maybe, my investment has an effect after all, and it’s not just the hidden growth potential.

37 PhilRossonml April 17, 2017 at 6:41 pm

I see your page needs some unique & fresh content.

Writing manually is time consuming, but there is solution for this
hard task. Just search for: Miftolo’s tools rewriter

Comments on this entry are closed.

Previous post:

Next post: