In February, when Covid-19 was mainly a Chinese phenomenon, that led to the lockdown of Wuhan and Hebei Province, most economists forecast that the virus would have little impact on global growth and that China's economy would slump in 1Q19 but snap back in V-shaped recovery in 2Q19. However, as the virus spread exponentially across the globe in March, economists began to take it seriously and to take down their growth forecasts and consider the possibilities of U-shaped or even L-shaped recoveries. By mid-month, some economists declared a global recession. By month-end, a debate was developing about whether it would be a Recession or a Depression.
For their part, Canadian economists began adjusting their forecasts in early March and progressively downgraded the growth outlook as the month wore on. The chart below shows how forecasts for Canada's real GDP have changed since late February.
The Chart shows real GDP in level terms; it was C$2.1 trillion seasonally adjusted at annual rates in 4Q19. On February 21, the economists at my old employer, JP Morgan, had a close-to-consensus forecast that the economy would grow steadily, if not robustly, at about its' potential rate of 1.7%. By March 13, JPM had downgraded the forecast to show a moderate recession, with growth contracting at annual rates (ar) of -1.5% in 1Q20 and -2.5 in 2Q20, followed by a rebound in growth to about a 3%ar in the second half of the year.
Two weeks later, on March 27, JPM had reassessed the damage and forecast contractions of -5.5%ar and -18.5%ar in the first two quarters of the year, followed by growth averaging 9%ar in the second half. For comparison, Bank of Montreal (BMO) Economics on March 27 forecast contractions of -6.5% and -25%ar in 1Q and 2Q20, followed by a rebound at a stunning 30%ar in 3Q20 and 4% in 4Q. TD Economics made a similar forecast on March 25, but didn't expect as fast a rebound in growth as BMO.
The convention of stating growth at annual rates confuses people by exaggerating the weakening of the economy. The reality, for those who are not growth afficionados, is that BMO, for example, is forecasting that seasonally-adjusted real GDP will contract by -1.7% in 1Q20 and by -7% in 2Q, followed by a 6.8% rebound in 3Q20 and 1% growth in 4Q20. This results in a full year contraction of real GDP of -3% in 2020. Forecasts for 2021 growth now stand at 2.7% for JPM, 3.5% for BMO and 3.7% for TD. It would be helpful and constructive for economists to stop reporting their quarterly forecasts at seasonally adjusted annual rates and just report the actual contraction of real GDP.
This is not to minimize the hit to the economy, it is just to clarify what these economists are telling us. In fact, their arithmetic points to a very sudden and serious recession. The range of forecasts is fairly large, even though all of the forecasters expect the economy to be recovering by 3Q20. All of the forecasters are assuming that Covid19 cases will peak in 2Q and subside in a manner similar to that experienced in China or South Korea. If new cases take longer to peak or if there is a resurgence of new infections in the autumn, the rebound in real GDP will be slower and take longer than in the current forecasts.
Even if the current forecasts are close to accurate, they point to a significant loss of real output and incomes. The output gap is forecast to widen to the 7-9% of GDP range in 2Q20. Real incomes will likely fall by even more as Canada's terms of trade have fallen sharply due to the drop in crude oil and other commodity prices. Even in the most optimistic forecast, real GDP will not return to its' potential by the end of 2021. BMO forecasts that even six quarters into the recovery, the output gap would still be almost 3% of GDP, while JPM and TD see the gap still at almost 4% of GDP. Forecasters seem to be suggesting that there Covid19 is causing a permanent loss of output and income and a downshift in the future path of the level of real GDP.
While economists have spent considerable effort attempting to estimate the immediate hit to real GDP, they seem to have spent less time thinking about the impact of Covid19 on inflation. The chart below shows some of the latest inflation forecasts.
Before significant adjustments in forecasts were made, as illustrated by JP Morgan's February 21 forecast, the outlook was for CPI inflation to remain well behaved and to converge to the Bank of Canada's 2% target by the end of 2020 and to remain there through 2021. By March 27, all forecasters had built in a dip in inflation in 2Q20 followed by a steady return to the 2% target or slightly higher by the end of 2021. BMO projects the biggest dip in inflation to just 0.3% over a year ago in 2Q20, and despite their forecast of a sharp snapback in real GDP in 3Q20, still expects inflation to be just over 1%oya in 1Q21. Perhaps because they have been well trained by the Bank of Canada's projections, all forecasters expect inflation to converge to close to target within two years. Yet, as mentioned above, these forecasters still expect an output gap (excess capacity) of between 3 and 4% of GDP by the end of 2021.
Some forecasters may rationalize this return to target inflation despite a still sizable output gap by assuming that potential GDP estimates will be permanently reduced by the Covid19 crisis. It is not clear why this should be the case. The growth of the labour force is not likely to be affected unless the government indefinitely closes the border to immigration. And there is no reason to mark down potential estimates of productivity unless economists believe that Covid19 will a permanent depressing effect on capital investment. These outcomes are possible, but only likely if the economic shutdown lasts much longer than is assumed in current forecasts.
Bank of Canada Governor Poloz noted the lowered growth and inflation forecasts as the Bank cut its policy rate to the "effective lower bound" of 0.25% from 1.75% at the end of February. Poloz said that the lowered forecasts he was seeing were little more than arithmetic and promised to unveil new projections on April 15th from the Bank of Canada's forecasting team, which he described as "the best there is". He suggested that more than just doing arithmetic, the Bank of Canada's forecasters would take account of confidence effects and how behavior might change in the post-Covid19 world. One suspects that the Bank will provide a range of scenarios based on differing assumptions about the course of the Covid19 pandemic and the likely economic repercussions of a shorter or longer crisis.
Many questions are not answered, or even addressed, in these forecast revisions. Will government support programs and a huge increase in deficit spending prevent corporate defaults and a permanent destruction of economic capacity. If debt defaults are too large, will that tigger persistent higher unemployment and unleash deflationary forces? Will Canada's highly productive energy industry be able to recover when government policies are tilted toward "phasing out" fossil fuels? Will financial markets take in stride the coming huge spike in Canada's federal debt-to-GDP ratio, or will investors demand a higher credit risk premium for a country which already has very high levels of household and corporate debt? Will the Bank of Canada's policy interest rate setting stay at the effective lower bound indefinitely, and if so, will bond markets anticipate higher inflation and bond yields? Perhaps the Bank of Canada's crack forecasting team will shed some light on these questions. But probably not.
No comments:
Post a Comment