Reconciling the Bank of Canada's forecast or the Importance of Exchange Rate Assumptions

In its most recent MPR, the Bank of Canada produced a growth forecast for 2010 that seems rather optimistic compared to some private sector forecasts:

Bank of Canada 3%
Merrill Lynch 2.7%
Bank of Nova Scotia 2.5%
Royal Bank of Canada 2.5%
Shock Minus Control 2.5%
Bank of Montreal 1.8%
CIBC 1.5%
Toronto-Dominion Bank 1.4%

One probable reason for the variance between my forecast and the Bank’s is the Bank's assumption of an 87 cent US/CAD exchange rate versus my model range of 87 to 95 cents. Indeed, I can account for quite a bit of the variance between my own forecast and the Bank's simply by tweaking the exchange rate assumption.

Of course, there any number or assumptions that could be driving this difference, particularly within a complex model such as TOTEM, the Bank’s primary forecasting model. However, the low exchange rate assumption does seem to be a pretty important one.

Canadian Economic Forecast 2009-2010

The Canadian economy will contract significantly in the first half of 2009 before giving way to a modest recovery in the third quarter. I anticipate that growth in 2010 will be given a boost from combined fiscal and monetary stimulus, but that overall economic growth in 2010 will be sluggish. Overall, I expect the Canadian economy to contract by 2.5% in 2009 before returning to positive growth of 2.4% in 2010. It is important to note that although I've forecast a strong rebound in quarterly real GDP growth in 2010, much of this growth will come from temporary sources such as government stimulus and the rebuilding of depleted inventories by Canadian businesses. Therefore, despite a return to positive quarterly growth, I expect that the economy will remain fundamentally weak, characterized by sluggish employment growth and an over reliance on government stimulus.

I expect the unemployment rate to peak around 9.2% before declining at a measured pace as the economy begins a slow recovery. The enormous amount of slack in the economy should keep core inflation well below the Bank of Canada’s 2% target for all of 2009 and 2010. Low inflation, along with a strong Canadian dollar, will allow the Bank of Canada to keep its conditional promise of holding its overnight rate at 25bps until the second quarter of 2010. However, if core inflation remains near target, the Bank may be eager to bring rates back to pre-crisis levels. The forecast suggest that the second half of 2010 could see the Bank increasing its overnight target by 100 basis points with the first increases announced in June or July.

I should note that this forecast assumes that the normalization of credit markets currently underway continues without further disruption and that GDP growth in the United States turns positive by the end of 2009 and remains positive but below trend growth in 2010. The forecast also assumes that the Canadian dollar stays within a range of 87-95 cents.

The recession is over!!...now about that recovery...

Canada has had three severe recessions in the past 30 years; one in 1981, again in 1990 and the most recent one that began in the last quarter of 2008.

The 1981 recession was a classic “V” shaped recession and lasted roughly five quarters. It was also the second part of a double-dip recession, following a brief contraction in 1980. The recession of 1990 was a steep, extremely painful and protracted “U/L” shaped recession, and was followed by a very weak recovery.

The Bank of Canada made some noise this past week after the release of its latest Monetary Policy Report (MPR) predicting an end to the recession in Q3 of this year. If the Bank’s forecast is correct (remember that it is a forecast, not an official statistical release) then the 2008 recession would go into the books as the shortest of the past three Canadian recessions.

The following figure illustrates the path of the Canadian economy 12 quarters after the business cycle peak prior to the recession for 1981 and 1990 along with the recent Bank of Canada forecast.



So where does the BoC expect the recovery to come from? Three places – Personal Consumption Expenditures, Government and Inventories. See below from the July MPR

I have my doubts that PCE will really be that strong coming out of this recession. THe unemployment rate is liklely to approach 10% and household saving is rising in response to massive wealth destruction from falling equity and home prices over the past year. However, indicators such as retail sales and consumer confidence do seem to be improving, so perhaps the Bank is right.

The real difference maker in the expected recovery from the 2008 recession, compared with past recessions, is the fiscal policy response of the Canadian Government. Government spending in the year following the end of the 1981 recession contributed just 0.3% to real GDP growth and only 0.2% to real GDP growth in the year following the end of the 1990 recession. In contrast, the BoC estimates that Government spending will add 1.3% to growth in 2010.

Given that inventory rebuild is estimated by the Bank to add 1% to growth in 2010 and you have 2.3% of the 3% growth forecast estimated to come from temporary sources that may not do much for job creation. Add an uncertain outlook for the Canadian housing and non-residential construction sector and it would seem like we have a recipe for a jobless recovery in 2010. But at least the recession is over...right?

Whats going on with the yield curve part 2: Does a steepening of slope imply growth?

I'm having a hard time with this one. The model I use for forecasting real GDP growth relies on changes in the slope of the yield curve as the mechanism by which monetary policy works. As discussed previously, the slope of the Government yield curve (10yr - 3month) has steepened dramatically in the past month. This steepening is often interpreted as a positive signal that market participants expect short-term rates to rise in the future as growth and eventually inflation pick-up. What I am having a difficult time with is whether this is the right way to interpret what is currently happening in the Canadian economy. The Bank of Canada is certainly not going to be raising rates any time soon, and research by the IMF shows that recoveries from recessions caused by financial crises tend to be slow.

The forecast implications are significant. Ignoring the signal from the yield curve implies a less robust recovery while accepting it provides a recovery akin to what the Bank of Canada forecast in its last MPR. The figure below illustrates the incremental growth implied by the yield curve shock (versus a control baseline of ignoring the signal, eg, running a simulation in which no steepening occurred

I'm leaning towards interpreting the movement in the yield curve as simply the normalization of inflation expectations and the impact of the "flight from quality" as investors regain confidence. Anyone want to try to convince me otherwise?

GDP better than expected in Q1

Real GDP growth in Canada came in at -5.4% (annualized) for the first quarter of 2009. It turns out that my forecast of -8.2% was off by a wide margin. Thankfully, I had prestigious company with David Wolf (Merrill Lynch), Marc Carney and the Parliamentary Budget Office all missing substantially on the low side. The figure below shows how some prominent (and not so prominent) forecasters performed for Q1 Real GDP growth - I'll spoil the surprise - we still (mostly) suck at forecasting GDP.

I haven't had time to delve into the details, but much of the surprise seems to have come from higher than expected consumer spending. Personal consumption expenditures were actually less of a drag on growth in Q12009 than in Q42008. Investment was atrocious, as expected and growth was helped out a little by imports falling more than exports.

A lot going on in the economy these days - loonie above 90 cents, 10-year yields rising rapidly -which means I'll be updating my forecast this week. Stay tuned.

Forecast update

I've updated the Canadian quarterly forecast ( see link in upper right corner) to reflect the weakness of incoming data. I've jumped on board with David Wolf and Kevin Page, forecasting a contraction in Q1 of 8.2%. I also expect Q2 to be significantly worse at -3.8% before the economy begins a recovery sometime in Q3-Q4 as a result of fiscal and monetary stimulus. This forecast is considerably more bearish than the Bay Street consensus (a good thing?).

A Bridge to Where?

Canadian economic forecasters have recently been drastically lowering their expectations for Q1 GDP from simply a very bad quarter, say around -4% to -5%, to expecting one of the worst quarters in Canadian history. So what gives? Is incoming data really that bad? How can one reconcile current data with quarterly forecasts?

Well, it turns out that there is a fair amount of literature on updating quarterly forecasts with monthly information. One technique that caught my eye (because of its simplicity) is the construction of a so-called “bridge” equation. A bridge equation is basically a simple regression of monthly economic variables, aggregated to quarterly values and then used to forecast quarterly GDP. An example of such an equation can be found in this Bank of Canada working paper.

The model in the BoC paper includes the consumer confidence index, hours worked, retail sales, housing starts, 3-quarter lagged Canadian GDP and US industrial production as explanatory variables. Based on the most recent data for those variables, the bridge model predicts Q1 real GDP growth of -8.3%. This compares with a quarterly model prediction of -4.4%.

So, using the bridging equation we end up at the same bad place arrived at by Bay Street and others– a really really really bad Q1.

This is bad news

Kevin Page of the PBO has joined David Wolf in forecasting a historic contraction in Q1:

"Mr. Page told the House of Commons finance committee that, based on private-sector forecasts and his own assessments, he expected GDP to contract by about 8.5% in the first quarter of 2009 and by 3.5% in the second quarter." - Globe and Mail

I've only recently begun trying to forecast GDP and in the model that I'm using, it would be very difficult to generate a contraction of that magnitude. I would love to see the details or the rational for the PBO and Merrill forecasts .

Merril Lynch forecasting -9.1% GDP contraction in Q1/2009

Wow - makes my prediction of -4.3% look like a call to rejoice . Not sure how ML arrived at that rather staggering number but lets hope David Wolf was just trying to get some publicity because if he's right, it will mean a lot more pain for Canadians.

Forecast Update

I've updated my 2009-2010 forecast to incorporate actuals from Q4/2008. Not much has changed, I've lowered the forecast for growth slightly and increased the expected unemployment rate. It remains a very difficult forecasting environment, especially in the sort of monetary policy driven model that I'm employing.

I'll benchmark my forecasts to the pros on Bay Street once Q1 GDP is released, that should be interesting.

(click to enlarge image)

2009 Revision and 2010 Forecast

Posted below is my 8 quarter forecast slightly revised for Q4 economic data and some model tweaks. I still expect 2009 to be quite weak, with average year-over-year growth of -1.3%. Also of note, I am now forecasting 4 quarters of negative growth from 2008Q4 to 2009Q3. If all goes well, the 2009 recession will be followed by a healthy recovery in 2010 with real GDP growth of 2.2% for the year, and over 3% in the second half. This 2010 forecast puts me more in the optimistic Bank of Canada camp than those pessimists at the IMF.

(Click to enlarge)

Canadian Quarterly Economic Forecast 2009

One of my objectives in starting this blog was to keep a record of my own economic forecasts (with the recognition that economists have a fairly pitiful forecasting record). Moreover, I haven't been quite satisfied with my initial crack at forecasting the Canadian economy in 2009. Here I present a revised quarterly forecast (though I'll stand by my original numbers for the contest).

As shown in the table, I am forecasting that the economy will be in recession for at least 3 quarters, from 2008Q4 to 2009Q2 with a real possibility of the recession extending into 2009Q3. Importantly, this forecast assumes government stimulus of 0.5% of GDP in 2009Q3 and 0.5% in 2009Q4 - that is, a stimulus package proposed in the January budget will take several months to impact the economy. This is an entirely arbitrary assumption, but one I can live with. Based on the most common number thrown around in the media I've assumed total stimulus of 2% of GDP, spread equally across four quarters (this is almost certainly wrong but maybe close enough). Without this stimulus, I would project that the recession would last through 2009Q4.

Without further ado, I present the inaugural Shock Minus Control forecast:












So how does my projected 2009 recession compare with past Canadian recessions? If my forecast is close, it should be fairly deep but perhaps not as drawn out as 1981 or 1990: