On 19th Century Tariffs & Growth

A few days ago, Pseudoerasmus published a blog piece on Bairoch’s argument that in the 19th century, the countries that had high tariffs also had fast growth.  It is a good piece that summarizes the litterature very well. However, there are some points that Pseudoerasmus eschews that are crucial to assessing the proper role of tariffs on growth. Most of these issues are related to data quality, but one may be the result of poor specification bias. For most of my comments, I will concentrate on Canada. This is because I know Canada best and that it features prominently in the literature for the 19th century as a case where protection did lead to growth. I am unconvinced for many reasons which will be seen below.

Data Quality

Here I will refrain my comments to the Canadian data which I know best. Of all the countries with available income data for the late 19th century, Canada is one of those with the richest data (alongside the UK, US and Australia). This is largely thanks to the work of M.C. Urquhart who recreated the Canadian GNP series fom 1870 to 1926 in collaborative effort with scholars like Marvin McInnis, Frank Lewis, Marion Steele and others.

However, even that data has flaws. For example, me and Michael Hinton have recomputed the GDP deflator to account for the fact that its consumption prices component did not include clothing. Since clothing prices behaved differently than the other prices from 1870 to 1885, this changes the level and trend of Canadian incomes per capita (this paper will be completed this winter, Michael is putting the finishing touch and its his baby).  However, like Morris Altman, our corrections indicate a faster rate of growth for Canada from 1870 to 1913, but in a different manner. For example, there is more growth than believed in the 1870-1879 period (before the introduction of the National Policy which increased protection) and more growth in the 1890-1913 period (the period of the wheat boom and of easing of trade restrictions).

Moreover, the work of Marrilyn Gerriets, Alex Chernoff, Kris Inwood and Jim Irwin (here, here, here, here) that we have a poor image of output in the Atlantic region – the region that would have been adversely affected by protectionism. Basically, the belief is a proper accounting of incomes in the Atlantic provinces would show lower levels and trends that would – at the national aggregated level – alter the pattern of growth.

I also believe that, for Quebec, there are metrological issues in the reporting of agricultural output. The French-Canadians tended to report volume units in manners poorly understood by enumerators but that these units were larger than the Non-French units. However, as time passed, census enumerators caught on and got the measures and corrections right. However, that means that agricultural output from French-Canadians was higher than reported in the earlier census but that it was more accurate in the later censuses. This error will lead to estimating more growth than what actually took place. (I have a paper on this issue that was given a revise and resubmit from Agricultural History). 

Take all of these measurements issue and you have enough doubt in the data underlying the methods that one should feel the need to be careful. In fact, if the sum of these (overall) minor flaws is sufficient to warrant caution, what does it say about Italian, Spanish, Portugese, French, Belgian, Irish or German GDP ( I am not saying they are bad, I am saying that I find Canada’s series to be better in relative terms).

How to measure protection?

The second issue is how to measure protection. Clemens and Williamson offered a measure of import duties revenue over imports volume. That is a shortcut that can be used when it is hard to measure effective protection. But, it may be a dangerous shortcut depending on the structure of protection.

Imagine that I set an import duty so high as to eliminate all entry of the good taxed (like Canada’s 300% import tax on butter today). At that level, there is zero revenue from butter import and zero imports of butter. Thus, the ratio of protection is … zero. But in reality, its a very restrictive regime that is not being measured.

More recent estimates for Canada produced by Ian Keay and Eugene Beaulieu (in separate papers, but Keay’s paper was a conference paper) attempted to measure more accurate indicators of protection and the burden imposed on Canadians. Beaulieu and his co-author found that using a better measure, Canada’s trade policy was 11% more restrictive than believed. Moreover, they found that the welfare loss kept increasing from 1870 to 1890 – reaching a figure equal to roughly 1.5% of GDP (a non-negligible social cost).

It ought to be noted though that alongside Lewis and Harris, Keay has found that the infant industry argument seems to apply to Canada (I am not convinced, notably for the reasons above regarding GDP measurements). However, that was in the case of Canada only and it could have been a simple outlier. Would the argument hold if better trade restriction measures were gathered for all other countries, thus making Canada into a weird exception?

James Buchanan to the rescue

My last argument is about political economy. Was the institutional arrangement of protection a way to curtail government growth? Protection is both a method for helping national industries and for raising revenues. However, the government cannot overprotect at the risk of loosing revenues. It must protect just enough to allow goods to continue entering to earn revenues from imports.  This tension is crucial especially since most 19th century countries did not have uniform general tariffs (like a flat 5% import duty) which would have very wide bases. The duties tended to concern a few goods very heavily relative to other goods. This means very narrow tax bases.

Standard public finance theory mandates wide tax bases with a focus on inelastic sources. However, someone with a public choice perspective (like James Buchanan) will argue that this offers the possibility for the government to grow. Basically, a public choice theorist will argue that the standard public finance viewpoint is that the sheep is tame. Self-interested politicians will exploit this tameness to be elected and this might imply growing government. However, with a narrow and elastic tax base, politicians are heavily constrained. In such a case, governments cannot grow as much.

The protection of the 19th century – identified by many as a source of growth – may thus simply be the symptom of an institutionnal arrangement that was meant to keep governments small. This may have stimulated growth by keeping other sectors of the economy more or less free of government meddling. So, maybe protection was the offspring of the least flawed institutional arrangement that could be adopted given the political economy of the time.

This last argument is the one that I find the most convincing in rebuttal to the Bairoch argument. It means that we are suffering from a poor specification bias: we have identified a symptom of something else as the cause of growth.

A Note on the Econometric Evaluation of Presidents

Sometimes, I feel that some authors simply evolve separately from all those who might be critical of their opinions. I feel that this hurts the discipline of economics since it is better to confront potentially discomforting opinions. And discomforting opinions are never found in intellectually homogeneous groups. However, a recent paper in the American Economic Review by Alan Blinder and Mark Watson suffers exactly from this issue.

Now, don’t get me wrong, the article is highly interesting and provides numerous factoids worth considering when debating economic policy and politics. Basically, the article considers the differences in economic performance under different presidents (and their party affiliation). Overall, it seems that Democrats have a slight edge – but in large part because of “luck” (roughly speaking).

However, no where in the list of references do we find an article to the public choice theory literature. And its not as if that field had nothing to say. There are tons of papers on policy decisions and the form of government. In the AER paper, this can be best seen when Blinder and Watson ask if it was Congress, instead of the president, that caused the differences in performance. That is a correct robustness check, but it is still a mis-specification. There is a strong literature on “divided government” in the field of public choice.

In the case of the United States, this would be presidents and congresses (or even different chambers of congress) of different party affiliation. Generally, government spending is found to grow much more slowly (even relative to GDP) when congress and the White House are held by different parties. Why not extend that conclusion to economic growth? I would not be surprised that lagged values of divided government (mixed partisanships in t minus one) would have a positive on non-lagged growth rates (growth in t-zero).

Now, this criticism is not sufficient to render uninteresting the Blinder-Watson paper. However, it shows that some points fall flat when two fields fail to link together. Public choice theory, in spite of the wide fame of James Buchanan (Nobel 1986), Gordon Tullock and affiliates (or off-spawns) like Elinor Ostrom (Nobel 2009), is still clearly unknown to some in the mainstream.

And that is a disappointment…

What is entrepreneurship and why should you care?

The word entrepreneurship is thrown around a lot, but rarely defined. As far as I can tell nobody really believes (or is willing to admit they believe) that entrepreneurship is anything less than highly important (“Green child entrepreneurs are our future! Support our troops against breast cancer!”). It’s probably a wise political move to not pin down the idea because it means anyone’s cronies can be considered entrepreneurs. Speaking of which, “crony” is almost the antonym of entrepreneur. Cronyism evokes images of stagnation, inefficiency, “innovations” that make things worse, and opportunities for genuine improvement that are ignored.

Invisible hand

So what does entrepreneurship mean? There are two general definitions, and both bear on the question of how to go about having a peaceful, productive, and morally praiseworthy society. The more general of the two is judgment in the face of uncertainty. That is, given that we don’t know what tomorrow will look like, and we certainly don’t know what the world will look like in 10, 25, or 50 years, we have to make wise, forward-looking decisions. We don’t have enough information to simply plug the relevant data into an Excel spreadsheet and get the “correct” action from a formula. In other words, understanding the ubiquity of entrepreneurship means that we still consider The Use of Knowledge in Society to be relevant.

The more specific definition is pursuit of pure economic profit (above “normal returns to capital, labor, etc.”) by pursuing hitherto un- or under-exploited opportunities. Such breaks from the status quo are the creative acts necessary for economic progress. Entrepreneurship is the human face of economic change that provides a micro-level description of what economists might otherwise wave their hands over and call “technology.” This sort of entrepreneurship is an important source of uncertainty about the future. 2014 is so much different from 1964 because of the actions of innovative entrepreneurs out to improve their own lives.

You’ll notice that I haven’t defined entrepreneurship in a way that actually is inimical to cronyism. That’s because not all entrepreneurship is productive. Destructive entrepreneurship is the pursuit of economic profit that makes the entrepreneur better off at the expense of someone else resulting in a net-loss. So we should be concerned not only with allowing individuals the autonomy necessary to be entrepreneurial (rather than merely reacting formulaically to top-down commands), but also with establishing institutions that direct people to help others.

Invisible hand

From the Comments: American Politics without the Romance

Longtime reader (and prolific blogger in his own right) –Rick riffs off of the Obama administration’s latest attempt to flaunt the rule of law:

Many of the tactics being used by Democrats and President Obama, today, derive from past tactics approved by Republican majorities and Republican Presidents who sought to avoid the difficult role of governing properly by seeking work around exceptions to the Constitution by reassigning or allowing the usurpation of powers between branches or though Constitutional amendments on requirements that stood as roadblocks.

So, neither party is better or more moral than the other in this regard.

Read the whole thing. Upon second thought, I probably should have titled this post “politics without romance” and just omitted the “American” part of it. In fact, you can pretty much use –Rick’s comment to explain every social conflict imaginable if you just make sure that the words ‘Democrat’ and ‘Republican’ are replaceable by any faction and the word ‘Constitution’ is interchangeable with the word ‘power.’

I got the phrase “politics without romance,” by the way, from Nobel laureate James Buchanan.

Has Foreign Affairs Been Reading NOL?

Hello all, I signed up for a pretty challenging final quarter here at school, so my postings will probably be scarce for the next two or three months. It seems Foreign Affairs, one of the more sober foreign policy journals out there, is finally starting to read us here at the consortium. I’ll get to that in a minute but first: editorial duties call!

  1. Be sure to read Dr. Delacroix’s Bush-worshiping piece for an example of how obstinate ignorance works. The very man who mocks smart, well-educated people for their acceptance of scientific consensus on global warming as ‘cultists‘ seems to believe that “there were very good reasons for any reasonable person to be misled about the existence of  [WMDs] in Iraq.” You have to admit, the man has a lot of brass!
  2. I still have to get to co-blogger Andrew Roth’s recent comment chastising conservatives and libertarians for failing to recognize the many nuances associated with Bismark’s statecraft and Roosevelt’s New Deal.
  3. We’ve got a couple new writers who will be blogging here at the consortium. One is an economics major at UC Merced and the other is a Guatemalan national doing graduate studies in Denmark, so stay tuned!

Political scientists Roland Benedikter and Lucas Kaelin have a fascinating piece in Foreign Affairs focusing on the one bright spot in Europe these days: Switzerland. Libertarians who have read the political and legal works of Friedrich Hayek, Ludwig von Mises and James Buchanan will recognize the gist of the arguments right away. To summarize: small, democratic states are the best form of government available to man, given our vast shortcomings, and these small states are, in turn, much better off operating within vast free trade zones that do not hinder the small-scale democracy at work in these states. From the piece: Continue reading

Property Rights in Africa: More Decentralization Please

From the economist Camilla Toulmin:

While land registration is often proposed as a means of resolving disputes, the introduction of central registration systems may actually exacerbate them. Elite groups may seek to assert claims over land which was not theirs under customary law, leaving local people to find that the land they thought was theirs has been registered to someone else. The high costs of registration, in money, time, and transport, make smallholders particularly vulnerable to this.

You can read the rest of her article here [ungated version can be found here]. It goes on to elaborate upon how more decentralization is needed, as well as the need for more incorporation of indigenous legal practices. Highly recommended, but grab a cup of coffee first.

Arguments to ponder:

  1. James Buchanan’s work on public choice (elite groups seeking to capture the rent)
  2. Friedrich Hayek’s work on tacit knowledge and the inability to plan societies from the top
  3. Elinor Ostrom’s work on governing the commons and how states muddle the intricate “rules of the game”

Any thoughts? Suggestions for further reading?