Recent Posts

Wednesday, May 29, 2019

Regional Recessions And Fiscal Policy

Regional economic divergences add to the complexity of recession dating. The fact that economic conditions vary across the country also complicates the policy response to recessions. The emphasis on automatic fiscal stabilisers in Modern Monetary Theory reflects this reality.

(This article is an unedited excerpt from my manuscript on recessions.)

Chart: Canada versus Alberta Unemployment Rate

The figure above demonstrates a recent example of diverging fortunes within the Canadian economy. The top panel shows the unemployment rate for 15- to 64-year old Canadians and for those living in the Province of Alberta. Until 2015, the Albertan unemployment rate moved in tandem with the Canada-wide rate, although at a lower level. However, the weakness in energy export prices after 2015 (bottom panel) hit the Albertan economy hard. (Albertan crude oil production is now dominated by the output of the tar sands, and those prices were notably weak versus globally traded oil prices due to the fact that there are only limited options for exporting the products.) The rise in the Albertan unemployment rate was comparable to that of 2008 crisis. Meanwhile, the Canadian unemployment rate only nudged higher.

If for someone based in Alberta, the fact that the national economy avoided recession could be viewed as a piece of trivia. Business plans are still going to be disrupted by lost jobs among customers.

Market analysts have tended to exaggerate the importance of commodities for the Canadian economy. If an analyst correctly predicted the fall in Canadian oil export prices, they could have easily predicted a recession in the Canadian economy. They would have been correct for the situation in Alberta, but it would have been a miss at the national level.

The existence of these divergences is one reason I am not particularly concerned about the details of recession-dating procedures. A national economy is the sum of regional economies. If national gross domestic product is on the verge of contraction, some regions are almost certainly undergoing economic contractions. (The only way that could be avoided is the low probability event that all regions have the exact same growth rate.) We need to ask why we are interested in determination of recession dates in the first place, as the side effects will likely be in evidence in some regions of the country.

If our concern is more financial crises, then these regional divergences are of less importance – putting aside the important exception of the euro zone. Generally, financial systems are largely national in scope, although the United States has a large regional banking system. Financial crises therefore tend to be national events, and there is a across-the-board tightening of credit conditions. (The Savings and Loan Crisis in the United States was somewhat regional in character.) The impact of the crisis dwarfs the discrepancies between the national regions.

(The euro area is more of a currency peg system than a true integrated economy. Financial systems are national, and financial crises can be contained within particular countries, as was the fate of the euro periphery after the Financial Crisis.)

Policy Implications

From a policy standpoint, regional divergences call into question the emphasis on aggregate demand management. The economic mainstream is fascinated by aggregated models, particularly those based on the notion of optimising agents. Interest rate policy is the preferred mode to deal with economic fluctuations, and interest rates are common across an entire currency bloc. To the extent that fiscal policy is accepted as a policy tool, it is thought of in terms of its effect on aggregate demand.
This stance fails when faced by regional divergences. The worst-case scenario is having some regions of the country contracting as a result of industrial weakness, while other regions are in the throes of a housing bubble. Interest rate cuts are unlikely to save a struggling industry – such as oil producers facing a collapse in the global oil price – while they will help a housing bubble inflate. (Housing markets will be discussed in greater length in {another chapter of the book}.)

Automatic fiscal stabilisers in the form of welfare payments, unemployment insurance, or the proposed Job Guarantee programme are one natural response to regional divergences. When we decompose the fiscal deficit on a regional basis, it will automatically react to regional disparities. That is, weaker regions will pay less tax, and citizens that benefit from those programmes will have their incomes boosted as a result of the transfers/wages.

Unfortunately, these automatic stabilisers are not enough by themselves to prevent recessions in an economy with after-tax wage disparities. For the automatic stabilisers to kick in, jobs have to be lost in the private sector – which is coincident with the start of the recession. (If one believed that expectations mattered greatly for economic outcomes, that might be sufficient to prevent recessions, as the expected effect of the automatic stabilisers could prevent self-fulfilling cutbacks in expenditures. However, the fact that recessions occur despite the existence of automatic stabilisers tells us that the belief that expectation management is enough to prevent recessions is implausible.)

Clouding Recession Calls

The developed economies have drifted into a slow growth regime with economic volatility. In such a regime, are likely to face many growth pauses with some regions dropping into contraction. Since my objective here is not to document every wiggle in activity, these regional recessions will not be pursued further. However, they will matter for market commentary, as well as for policy analysis.

(c) Brian Romanchuk 2019


  1. What do you think of the Cdn central bank taking some credit for that recession miss when commodity prices fell? I believe they lowered rates in early 2015 and they believe that helped a lot in avoiding a recession. These days however after "unlearning" a lot of economics, I'm not sure monetary policy is as powerful as textbooks suggest with its multiple channels (price of money, asset prices, exchange rate and "expectations").

    1. Not sure whether they can take sole credit; from what I recall, there was a decent amount of infrastructure spending. However, not raising rates matters for the over-extended housing market.

  2. This comment has been removed by a blog administrator.


Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.