What is the cost of an independent Canada?

In the last post, I discussed Canada’s mysterious productivity gap with the United States. It appears to have been rather long-lasting. The current explanation is that the gap derives the fact that Canada is a smaller market and firms therefore have trouble reaching the optimal size and specialization. The implication is that had Canada been part of the United States, it would be more productive today. 

But does the gap really exist? Might the whole post be based on a misconception?

Vincent Geloso suggested that the gap may be driven by Quebec, but that does not match the data. The income gap with the U.S. holds for Ontario back to the 1870s. Moreover, we have detailed firm-level data for 1871 which shows that Ontarian firms were about 3-5% less productive than their Michigan equivalents. That is not to say that Quebec wasn’t a laggard! It is to say that the gap, inasmuch as it exists, is not due to backwardness in Quebec or the Maritimes. 

Randy McDonald, however, suggested that we might be asking the wrong question. Maybe the question shouldn’t be “Why are Canadian firms less productive than American firms?” but rather “Are Ontarian firms less productive than firms in the Great Lakes states?”

After all, were Canada part of the United States, then modern Ontario would be the State of West Canada with a northern border somewhere between Lake Nipissing and the Albany River. West Canada would be just another Great Lakes state. If real genuine Ontario has been consistently less productive than the Great Lakes states, then we can conclude that there is some Canadian factor holding firms back. But if it is not, then we might want to doubt the entire premise.  

It is easy enough to get regional GDP per capita for 1997-2016. Now, GDP per capita is a highly imperfect proxy for productivity. It does not account for differences in age structure, labor force participation, or work hours. But it is what I have. The data for the U.S. comes from the BLS. The Ontarian data is converted to U.S. dollars using the OECD PPP exchange rate. 

And so:

Wow! Ontario was ahead of the Great Lakes states in the late 1990s, falling behind only after 2003. It fell behind Michigan only in 2010. This makes me wish for a longer time series. Is 1997 or 2016 the aberration? Has Ontario basically looked like a Great Lakes state since 1871? There is a paper for an enterprising graduate student here. 

In fact, there are two different axes upon which to compare real Ontario with counterfactual West Canada. The intensive axis is productivity. Would workers and firms be more productive in Ontario had they enjoyed the same access to the American market as firms in the Great Lakes states? Did the value of being American, if any, change over time? The question is interesting in-and-of itself, but it also might shed light on the putative benefits of European union.

But there is an extensive axis as well! Before 1879, the problem with being Canadian was that the United States imposed high tariffs on Canadian exports. After 1879, however, the National Policy also imposed very high tariffs on American and British exports to Canada. Canadian firms still lacked access to American markets but they no longer had to worry about American (or British) competition.

If the policy worked, then firms would be created in Canada that otherwise might not have existed. Those firms would require workers. Where would those workers come from? They could come from farms, but the Canadian farm sector was still growing fast. They could come from existing firms, but that would raise wages. In an age of open immigration, high wages should have attracted more immigrants. As the industrial heart of Canada, Ontario should have had more firms and more people than it would have had it been West Canada, just another Great Lakes state. (If I had to guess, I would imagine that Wisconsin might be the best parallel for counterfactual West Canada, despite its closer proximity to Michigan.) 

But did the National Policy work as advertised? And if so, how much bigger is Canada’s population? And what were its costs and benefits relative to (1) the reality of American and Canadian protection; (2) a counterfactual of American protection and Canadian openness (i.e., no National Policy); (3) a counterfactual of American and Canadian openness (i.e., annexation or a North American customs union); and (4) a counterfactual of American protectionism but Canadian openness only to Britain (i.e., some sort imperial federation)?

If anyone knows of a literature, please point me to it! And if there is not a literature, then somebody should get on it! Economic history, like most things, needs more Canadian content.

Comments

[J.H.]

"....then modern Ontario would be the State of West Canada with a northern border somewhere between Lake Nipissing and the Albany River"

Hmmm...I think you underestimate the size of the putative State of West Canada here. The nearest we thing to have to evidence as to how the US would likely have organized an annexed Canada is the 1866 Annexation Bill (see: https://en.wikipedia.org/wiki/Annexation_Bill_of_1866), which in Article IV (4) establishes "Canada West, with the addition of territory south of Hudson's bay and between longitude eighty degrees longitude ninety degrees;"

West Canada would have been closer in size to modern Ontario I think.

[Noel Maurer]

I was imagining either a different Treaty of 1783 or a victory in the War of 1812 (since the 1866 bill wasn't a serious proposal) but it doesn't really matter. Ontarian economic activity is almost entirely in southern Ontario.

A border at the Albany River insures the area north will remain a territory indefinitely. A border at the French river opens a slight possibility of a second state in the area, assuming that Sudbury sees the same growth and mining boom as it did in real life. I suspect that Thunder Bay would wind up in an expanded Minnesota or a separate Manitoba rather than in an oddly-shaped ... uh ... Rupertsland? Selkirk?

But again, given that pretty much all the industry is south of the French River, the boundary doesn't matter for the counterfactual.

[Vincent Geloso]

The point is well taken as it emerges from being unclear. I expressed myself very poorly. Three points were made in my original comment. The first is that I agreed with your explanation but questioned how large of a share does it explain. This is linked to my second point regarding Quebec. If we check Altman's recomputations for GDP per capita, the USA in 1891 are equal to 153 (where Canada=100) and within Canada, Ontario = 112 and BC = 162. Quebec = 85. Given that Quebec was then close to 30% of Canada's population, its role in explaining the lag is not-negligible. By 1911, Canada had closed the gap (USA = 113) but Quebec had fallen behind to 76. From my book, I show that between 1870 and 1940 - Quebec fell back in relative terms within Canada while Canada as a whole caught up with the US (see also Marc Egnal on this). That leads to my third point which is that we should really be talking about institutions. The first part is that Quebec had a less mobile population meaning that politicians were not as constrained by "moving with your feet". The other part is that free trade itself is an institution. The "great deal" at Confederation was that the Catholic Church in Quebec would ally with the Ontario protectionist (even if Quebec leaned free-trade) to give protectionism a boost in exchange for constitutional amendments that protected the Church's power (provincial control over social policy = education and health). As such, the key still lies in Quebec as a laggard but as an influencer of overall policy.

[Vincent Geloso]

The point is well taken as it emerges from ME being unclear. ***** (sorry for the typo)

[Noel Maurer]

The Altman figures I have from p. 163 of his 2012 book show U.S. per capita income falling from 153% of Canada to 124% in 1910, before rising again to 131% by 1929. Why the discrepancy with the figures cited above?

(BTW, I am going to order your book, although wow the pricing gives sticker shock!)

[Vincent Geloso]

Hmmmmmm, very good question especially since he took the time to calculate all of this in this 2003 AEHR article (http://onlinelibrary.wiley.com/doi/10.1046/j.1467-8446.2003.00053.x/full - page 250, panel B of table 2). For my book, send me an email and I'll get you something.

[shah8]

There have been quite a bit of writing about the Canada's management of a commodities-industry cycle. When commodities are high, industry is allowed to downswing a bit, and when low commodity prices provoke low loonie price, industry goes back on the upswing. One feature of the Harper Gov't was the breaking of this particular style in favor of an all-in oil boom.

[Randy McDonald]

It was suggested over on Facebook, where I shared this link, that Canada West might fare worse than Ontario simply because it's not going to be a national centre. How many companies will set up their headquarters in Toronto? Will there be anything like an Auto Pact to stimulate industrial growth? Will Toronto be a financial centre of any size? It may, instead, end up being peripheralized to the benefit of American cities.

(This is highly contingent, but what if the cities of Canada West end up sharing in the more negative trends of cities in the American part of the Great Lakes. A Toronto that evolves like Buffalo, say, never mind Detroit, is going to underperform.)

[J.H.]

That's a very good point Randy has there. I would imagine though that a similar but even stronger process would be at play with Ottawa versus Toronto as well as American cities. Ottawa may never reach any substantial size.

Given Toronto's role as a port city though perhaps Cleveland might be a slightly better model than Buffalo?

[Noel Maurer]

Hi, Randy,

That's an interesting point. Some things were lumped together, so let's break it down.

(1) The Auto Pact isn't relevant if Canada West is part of the United States. The Pact allowed Canadian plants to specialize, but if Canada already inside the U.S. market then those plants would already have been specialized.

It is possible that Canadian auto production might be lower /before/ 1965, but that's a subset of the discussion in the third-to-last paragraph of the post.

Am I missing something? If I'm not, please pass this on to your Facebook correspondents! I fear they may have missed the point.

(2) It's hard to say about corporate headquarters. It isn't like Midwestern cities are devoid of them. It is important to note that most Midwestern metro areas have done okay (Buffalo is an exception) over the past half-century.

(3) That's a great point about finance! Toronto employs roughly 250,000 people in finance; that's more than Chicago, and Chicago is no slouch. (Chart 18.) In Detroit, the number is 38,000. Toronto would certainly be smaller and possibly poorer, given that finance jobs were well paid. The jobs would still exist, of course, just elsewhere in North America.

[Randy McDonald]

To make things even more complicated, the existence of a Canada West whether bounded by the Albany or the French is hugely contingent on Loyalists being settled in Upper Canada in the 1780s, creating an Anglo settlement distinct from the French in lower Canada. Without that leapfrogging, settlement by Anglos would be postponed--Michigan was only settled in the 1820s, IIRC--and the area might well have been settled first by Canadiens coming up from the St. Lawrence.

If you do get an Upper Canada/Canada West/Ontario in the Union, well, what happened? Snapped up in the War of 1812, maybe?

[Noel Maurer]

Good point, again. The logic applies to a different War of 1812, which in my defense more plausible than a better outcome in 1783.

If Canada is part of US from 1783 onwards, then you’re right. I doubt that the Congress under the Articles would try to shoehorn Upper Canada into the Northwest Ordinance. It would all be just one big State of Canada, with all the resulting problems that Vincent pointed out.

[J.H.]

Another minor note is that if it was incorporated in a different war of 1812, would there be any motivation to change the name of the city from York to Toronto in the 1830s? From what I understand it was done through a petition of the residents, partly to move away from negative associations with the city as it was currently named and as some liked the name "Toronto" as the original name of the area. A different war of 1812 though probably means new settlers from the original US as well.

[Noel Maurer]

There might be some pressure to "Americanize" the name. And the town would likely have been sacked during the war regardless. So I'm not sure that we can make a systematic prediction about that.

What we can much more certainly predict is that the City of Toronto/York would find its municipal expansion stalling out with rather more suburban fragmentation than occurred in our Ontario. And there will be no amalgamation.

As a result, metro Toronto will be unrecognizable. It's population will likely be smaller and its per capita GDP will (before the 21st century, at least) be quite higher. But a small Toronto City will be hit by massive while flight during the Great Migration and likely look a lot like Cleveland or Milwaukee, even if Toronto County and the great metro area do fine.

In that sense, the people on Randy's facebook page are almost certainly right. Metro Toronto will be smaller but at least as prosperous if not more. But it will surround a severely-segregated and partially-abandoned inner city that will only be showing some faint signs of recovery in the last decade or so, utterly unlike the booming inner Toronto that we know.

[Noel Maurer]

I suppose Toronto City, WC, is more likely to be surrounded by "York County" rather than "Toronto County."

I suspect that Upper Canada would have been renamed "West Canada" rather than "Canada West"; there's something oddly British about the latter construction. But I don't know! Why did they pick the odd word order?

[J.H.]

I suppose they picked the odd word order (and it sounds oddly British) because prior to Confederation in 1867, the Province of Canada was divided into two administrative divisions/parts - "Canada West" and "Canada East". So "Canada West" was the actual name of the area under the British prior to Confederation.

As for the word order itself, I strongly suspect it has to do with the heavy French settlement and presence. After all the word order for the names of the predecessor entities to Ontario and Quebec followed French word order when translated into English:

Haut-Canada = Upper Canada
Bas-Canada = Lower Canada

Canada-Ouest = Canada West (not "West Canada")
Canada-Est = Canada East (not "East Canada")

[Randy McDonald]

I would just add that the State of Canada would evolve very differently. French Canada folded in on itself and became very conservative largely in response to British pressure at assimilation. If _Canadiens_ are free Francophone citizens of the State of Canada and the wider United States, well, I don't see where the push for an embrace of conservatism will come from. I doubt, for instance, that there would necessarily be such sustained underinvestment in human capital, education being made a domain of the church.

But, yes. A conquest of Upper Canada--not necessarily Lower Canada--in the War of 1812 is definitely imaginable. FWIW.

[Noel Maurer]

Assuming that the State of Canada becomes part of the United States in 1783, there are two possible outcomes.

In one, Canada relinquishes its claim on Upper Canada as part of the Northwest Ordinance. I find this unlikely, but it could happen.

In the other, Canada retains upper Canada, and faces a relatively large influx of Anglo settlers. It's a frontier, after all.

How would the Canadiens react to that? And what would the new settlers demand? The federal government is not going to intervene the way the British Empire did, but the local politics would be ... interesting. I don't know enough to opine how. Thoughts?

[Randy McDonald]

I honestly don't know. I am willing to make the bet that, absent the Loyalist settlement in Ontario, the advancing wave of Anglo settlement into the Great Lakes would have been preempted by Canadiens advancing up the Saint Lawrence valley. There was a thin but real tissue of French Canadian settlement upstream of Montréal, past the old French forts like Rouillé at the mouth of the Humber, up at least as far as Détroit.

To a certain extent, the relatively narrow boundaries drawn for Lower Canada led French Canadian migrants to opt for less attractive lands now outside their domain. The Laurentians and the Eastern Townships were settled substantially as a result of this, for instance. I can easily imagine the rich farmlands of the Lake Ontario, at least, attracting substantial numbers of migrants.

Then you get an influx of Anglos, coming into a frontier of a relatively established and decidedly non-Anglo society, with laws and customs different from their own. What happens? Well, I suppose Louisiana might be worth looking into, with the singular difference between that the Francophones of Louisiana would be much more riven by race and ethnicity than the Francophones of Canada ever were. Some degree of conflict is plausible, but how much?

[Noel Maurer]

I don't know either!

I should point out that I think I was wrong: it's by no means certain that a pre-1783 divergence would mean that the State of Canada would include Upper Canada. The province of New York claimed Upper Canada; another reason why the Quebec Act bit so much. Depending on the Canadiens, I think shoehorning Upper Canada into the Northwest Ordinance is more likely than we might think.

But if that didn't happen ... huh. What would the state government of Canada be like? There are certainly ways to discourage and deflect Anglo settlement. Presumably any royal landholdings would come under the control of Quebec City, not Washington, so the province could do what it wanted. But what would it do?

[Randy McDonald]

I guess maintaining a non-common law system, with an established state religion that is not Protestant, might do much to discourage settlement by Anglo migrants used to something different. Couple this with some sort of enduring political alliance between the Canadien political leaders who would presumably dominate affairs in the State of Canada and political factions elsewhere, and I think you could have something workable.

(FAN Quebec developed roughly along these lines, now that I think about it.)

I agree that, if we're talking about an American Canada, you could get any number of frontiers. What if the 45th parallel remained the Canadian southern border on the western shores of the St. Lawrence, too? What of the entire Ottawa river was linked to Canada? I don't think smaller boundaries would necessarily make much sense, if we're assuming an alliance between Canadiens and the Union, unless we're talking about territories in the Gulf of St. Lawrence (Magdalen Islands?) that Britain keeps.

Michigan and points beyond is implausible, I think we both agree, but Ontario may well be doable. You've got much more background in US history than me, so your call.

[Noel Maurer]

I don't know enough about Canadien political leadership or the weight they put on keeping Upper Canada to be able to guess. If they didn't care and weren't savvy, then the N.Y.-Quebec boundary is extended west, with the the area likely falling under the Northwest Ordinance and entering the union as a new state. Let's call it, I dunno, Ontario.

(The State of Canada West/West Canada is the name if the area is annexed around 1812; the State of Ontario is the name if it comes in around 1783 but gets hived off from the rest of Canada.)

But if the Canadiens were politically savvy and wanted Upper Canada, then I'm sure a deal could be made. New York's claims to the west were never more than aspirational ... but getting Canada into the Union was something the commercial elite very seriously wanted. So a deal could be struck.

The question, then, becomes: how fast would Anglos move into the territory? There was plenty of land elsewhere, so it wouldn't be terribly hard to discourage them. It didn't work in Louisiana, but plantation agriculture and the area's unique geographic advantages make the political economy a poor model for Canada State.

Slow migration and Canada State becomes a very distinct society, although I'm not equipped to speculate how. Fast migration and you get political conflict, including potential partion. Which did have some precedent: Vermont and Maine come to mind.

[Randy McDonald]

The only way I can imagine the Canadiens tossing in with the American rebels is a scenario where the British do not try to conciliate them via the Quebec Act, but rather decide to repress them in a manner not unfamiliar to their Irish co-religionists. That would probably be the only way an alliance with the Bostonnais would fly: if Britain is palpably worse, well.

This might be possible, although the British oppression of a sizable French colonial population will have knock-ons on relations with France. It would change things radically, though.

Until the construction of the Erie Canal or an analogue, the Canadiens have a natural advantage when it comes to expansion up the Great Lakes. 1820s, I think.

[J.H.]

But if the Canadiens aren't wooed with the Quebec Act, doesn't that remove one of the reasons for rebellion by the Anglo colonies in the first place?

[Randy McDonald]

Exactly. That alternate history starts skewing in interesting directions from the start.

[J.H.]

Seems more likely then that to still get a successful rebellion (one involving the 13 Anglo colonies at least, if not the other Anglo colonies too) instead of a Quebecois revolt that gets put down by the British (ironically with help from the Anglo colonies) it would take both the Quebec Act and a successful rebel conquest of Canada. That leads to Canada being brought in by 1783 at least and would likely lead to territory being cleaved off of the conquered province in recognition of the claims of the Anglo colonies (which in turn cede the areas as part of the Northwest Territory under the Northwest Ordinance).