The IMF this week released projections for the world economy over the next two years that were both optimistic and pessimistic.

When the projections were released, I messaged a fellow economics journalist and suggested they were pretty optimistic – he replied that he hoped they would be correct because they were predicting it would only be the worst downturn since the Great Depression.

Optimism comes in strange forms these days.

The positive outlook of the IMF is that, while they predict Australia’s GDP will fall by 6.7% this year (the worst fall since the Great Depression), they also predict it will rebound next year to rise by 6.1% (the best result since 1984 and second-strongest for 50 years).

That assessment suggested the IMF was predicting a “V-shaped” recovery.

There is no real science to any of this, but economists and commentators like to talk about recessions in terms of the shapes of letters.

“V” means basically an equally sharp fall and recovery, which is the most desired type (if you can desire a recession). A “W” shape is one where the quick recovery is hit with another slump and recovery.

“U” shape means the economy stays weak for a couple years, and then recovers strongly.

Finally there is the dreaded “L” shape, in which the economy stays weak for even longer – five to 10 years of barely enough growth to keep unemployment flat, but not enough to improve things all that much.

So for now the IMF (which gets advice from Treasury to formulate its estimates) sees a “V” shaped recovery. And yet not all is good.

While it expects the economy to grow almost as fast in 2021 as it did in 1984, unlike then, it expects the unemployment rate to rise – from 7.6% this year to 8.9% by the end of next year.

This is where the IMF is both optimistic and pessimistic.

There is a pretty solid relationship between GDP growth and unemployment (known as Okun’s Law) whereby the faster GDP grows the more the unemployment rate falls.

Over the past 50 years, we have needed the economy to grow by around 3.2% to keep the unemployment rate steady. In the past 20 years or so this has come down a bit, so now around 3% is good enough.

As a general rule, for every 1% that the economy grows above or below 3%, the unemployment rate will fall or rise 0.5%pts.

So, for example, given the current unemployment rate is 5.2%, if the economy was to grow by only 2% in a year, you would expect the unemployment rate to rise 0.5%pts to 5.7%.

Using the past 50 years as a guide, if the economy was to shrink by 6.7% in a year, as the IMF predicts, then the unemployment rate would be expected to rise by around 4.5%pts – which would take us up to 9.7%. And yet the IMF predicts unemployment this year to rise only to 7.6%.

That is optimistic and takes into account the government’s jobkeeper policy, which will see some people who would normally be classified as unemployed notionally still working.

But there the optimism ends.

While the IMF expects the economy to grow by 6.1% in 2021 – which would normally mean the unemployment rate should fall by around 1.5%pts – it predicts the unemployment rate to rise another 1.3%pts to 8.9%.

What this means is that while production might “snap back” (to use the prime minister’s phrase) we are likely in for a much tougher recovery on the employment front. Mining production may again ramp up, but labour-heavy sectors like tourism, hospitality and retail will be much slower to return to where we were just three months ago.

It means reducing government stimulus will also need to be done with care because while a “V” shape recession is hoped for, rushing the recovery and abruptly cutting stimulus can very easily see the “L” shape become a reality.

• Greg Jericho writes on economics for Guardian Australia