Does fiscal stimulus make sense in a supply-shock recession?

David Andolfatto is somewhat concerned at the lack of thought that's been put into justifying a Canadian fiscal stimulus program. This is a theme that both Nick and I have touched on over the past few months, and the analysis is complicated by the fact that too many commentators make the assumption that Canada and the US are facing similar problems and should therefore adopt similar policy responses. But Canada is not the United States, and it's far from clear that a rush to put infrastructure projects in place is a priority. (See also this post.)

More below the fold.

Firstly, we have to deal with the question of why Canada is in recession. The obvious place to start is the US recession, but that's at best a partial explanation. A US slowdown reduces Canadian aggregate demand, but not by enough to generate a severe recession.

What is enough to generate a severe recession appears to be a sudden drop in our terms of trade. To the extent that we'd try to shoehorn a terms of trade shock into a standard AS-AD model, it would probably do so as a technology shock. Trade can be conveniently thought of as a form of technology, so a negative terms of trade shock can be interpreted as a sudden reduction in our ability to transform commodities into consumer goods.

How should we react to a negative supply shock? If it were permanent, then there's not much for it except to get used to lower standards of living. Fortunately, there's very good reason to believe that this reduction is temporary; the long-run outlook for the prices of oil and other commodities is still consistent with a scenario in which Canada's the terms of trade return to their previous levels.

This isn't an argument for trying to avoid a deficit, of course. If you're faced with an explicitly temporary reduction in income, the optimal response is to borrow/dissave in order to keep spending levels more or less constant. In other words, it does make sense for the various levels of government to run deficits in order to provide income support over the short term. And since Canada's governments have had the good sense to run surpluses for the past dozen years, they can do so without worrying too much about hitting our debt wall. But this reasoning leads us to think of spending money on such things as expanding EI eligibility and on reinforcing other elements of the social safety net, not on infrastructure programs.

It would be an overstatement to claim that this is a pure supply-side recession, of course: the problems of the auto sector in particular are consistent with a story of a demand-side recession. The decline of the auto sector - and manufacturing in general - is part of a long-run trend, and it's a trend that has gone surprisingly smoothly up until now. But a managed decline is one thing; a catastrophic collapse is another. Indeed, I don't understand the point of making government aid conditional on a scenario in which the Detroit Three are viable in the long term. The long run will take care of itself; what we want to do is avoid large short-term employment losses. So I do see the point of providing public money here - although again, it's not a case for infrastructure spending, either. Unemployed auto workers won't be hired to build bridges and highways. Neither will unemployed drywallers, come to that.

Comments

[Phillip Huggan]

The counter argument would be now labour and materials are cheap for construction projects. City of Wpg road contruction projects were going over budget about 15-20% during commodity bubble. City of Winnipeg wished it has money for bike paths, a high speed transit corridor and other transit upgrades, but is unfortunately being forced to pick only one, for now (better transit cameras and shelters and scheduling), to give a weasely example.
The cost savings are maybe 25% in a recession? I guess you have to quantify whether the redirection of resources away from future booming resource extraction leads to natural resources extraction cost inflation or other costs that total more than the "stimulus" construction cost savings.
Stephen, I think an interesting line of research is why you are assuming a recovery. If this contraction happened after the USA election and/or if S.Palin could/would memorize responses to foreign policy questions, the USA response would be wealthy tax cuts which I'd guess would change your belief about a nearer-term recovery. I guess what I'm saying is at some point of inflection, it seems the stimulus response itself is either seen to psychologically, or genuinely does via increased demand from construction workers, initiate a recovery.
This might suggest Canadian overstimulating in a recession assuming non-Keynesian USA actor, and vice versa.

Ideally Zenn could've absorbed unemployed auto workers with subsidize public fleet demand; it is looking like the first mover advantage is already lost here. Ideally plants would be retooled to build wind turbines, but Ontario *clearly* wasn't for that outside of glorious Toronto. Now I'm wondering what autoworkers should do if not be sopped up by green jobs? I guess the plan is to hope for massive USA auto sector subsidies and keep them on pogey until then? Sounds risky under most historic USA Senate, House and Prez configurations. I wouldn't use them fluking out on a great Prez following all-time low approval rating Administration as the Canadian textbook model.

[Patrick]

There's some infrastructure we need anyway so we might as well build it while labor and materials are cheaper. Personally, I'd like to see more emphasis on rail and water transportation rather than roads.

Don't know if it counts as infrastructure spending, but at least in AB, it would be a good time to put money/effort into diversifying the economy away from oil and gas. We also need to figure out how to sequester oil & gas money during the booms to prevent inflation. During the boom the cost of materials, housing/houses, labor, food, etc was rising at a crazy rate. When oil prices go back into the stratosphere (as I think they inevitably will), AB will be right back into it's own special version of the Dutch disease. There will be too much money to be made in the oil patch for anyone to do anything else. The commodity boom will fuel another big rise in the value of the CDN$ and that will further discourage diversification.

[Alex Plante]

I'm a civil engineer who works for a large engineering consulting firm in the Montreal area. I design municipal infrastructure. Other colleagues design roads for the MTQ. I've noticed the following trends in the past few months:

- There has been an increase in infrastructure work for the MTQ and hydro utilities;
- There has been a moderate drop-off in municipal projects, but nothing too dramatic.

In the municipal engineering department, most of our work is for municipalities, but up to a quarter is for large building projects and electrical utility projects (for service connections, parking lots, etc.) Approx. 3/4 of the projects we do for municipalities are rehabilitation projects and feasibility studies, and only about a quarter are for new streets. I expect that as long as municipal finances are relatively sound we will be kept busy replacing, upgrading or rehabilitating existing infrastructure.

My personnal opinion is that since infrastructure is increadibly expensive and lasts such a long time that decisions to invest should be done on the basis of long-term planning, and not on short-term considerations such as the business cycle. Any infrastructure stimulus plan should focus on maintaining the current level of investment for projects that are economically justifiable, in order to avoid the collapse in investment which occured during the 80's and 90's, and for which we are now paying dearly in crumbling infrastructure and a shortage of experienced designers and builders.

[bob]

I can't see any way of characterizing the drop in demand for oil, and hence the change in terms of trade as a supply shock. That just seems totally backwards to me.

If it were a supply-shock recession, sure, fiscal stimulus wouldn't be the best response, but that is clearly not the case.

re: Alex Plante's point

My ideal situation would be to have the government keep a to-do list of big long-term projects, shelve those projects during the booms and wait to implement them during downturns.

[Adam P]

bob,

I think Stephen has in mind a story like the following.

Suppose that oil was discovered by the inhabitants of Gilligan's island, they pumped the oil from the ground and sold it on the world market. They bought all sorts of really cool stuff with the proceeds (but didn't leave the island that was their home). They have a stock of capital (drilling equipment, maybe some piplines) and they supply labour. Although we normally think of them as producing oil, through the magic of international trade they end up with all this cool stuff, not just oil. The net effect is they supply labour, employ capital and at the other end get a bunch of stuff. From their point of view, this system, including the international trade, is a production technology that produces all the cool stuff, to them the technology (taken as a whole) produces much more than oil.

Now suppose the real price of oil falls by half, to the people on the island the effect is this, for the same capital input and same labour input they get half as much cool stuff at the end. Another way to say it is:

For the same capital input and the same labour input they can produce half as much.

This last sentance certainly sounds like the definition of an adverse productivity shock. I imagine this is something like what Stephen had in mind.

[Stephen Gordon]

Yes, that's about it. For example, the *quantity* of oil that we're exporting has held up. The problem is the the price has fallen - and with it, our buying power.

[westslope]

I sure hope the terms of trade do not return to where they just were! (Nor do I expect them to.) Canada in general, western Canada in particular, is still labouring under a philosophy of get-rich-quick which means that much of the benefits would be squandered in the next gold rush.

Although I agree with the emphasis on reinforcing the social safety net, is the Employment Insurance (sic) program the best instrument for doing that? The EI, formerly Unemployment Insurance program, already acts as an income transfer program for seasonal resource workers, commercial fishing entrepreneurs, and pregnant mothers.

Or is polite face-saving BS simply part of who Canadians are? Regardless of the market distortions, the renewable resource value destruction, and Canada's status as hewers of wood and haulers of water that the EI program currently reinforces?

[reason]

Actually, the words "gold rush" are significant to me. Does anybody have a good economic model of a gold rush? A gold rush is a destructive sort of bubble, that makes a few much richer and most poorer in the intermediate term (but can have good long term consequences if it promotes immigration to areas that have something other than gold to offer). I think the US suffered from a financial gold rush in the last couple of decades and it is a good model to play with.

[reason]

Westslope,
yes another reason for promoting GMI.

[bob]

Adam P & Stephen -

Ah, thanks. I think I see what you are saying now. I was getting hung up on the supply shock term (thinking of OPEC in the 70s) and thinking in global terms. On the national scale, I guess it would be useful to look at it as a productivity shock, because the global decline in the oil price is outside our ability to influence, and wasn't caused by a decline in domestic demand.