If you are unemployed you are entitled to feel dreadful.

A year ago you were one among 686,000. Now you are among 789,000. An extra 103,000 Australians have joined you in the past year making your chances of being re-employed a good deal worse.

But, read on, it’s probably not as bad as you think.

First, politics. Why politics? Because Tony Abbott himself committed the Coalition to “creating 1 million new jobs within five years”. So far his record is awful. All through the global financial crisis, all through the Rudd and Gillard and Rudd years, all through the political crises, Australia’s unemployment rate never reached 6 per cent. It hit 6 in January and now it’s 6.4. Appallingly, Australia’s unemployment rate is now higher than that of the United States. We’ve passed the US on the way up as it’s on the way down.

Employment minister Eric Abetz has a ready excuse: “The carbon tax was removed just in recent weeks, and it stands to reason that you won't see an immediate overnight economic recovery,” he told the ABC’s PM program.

But an immediate improvement is exactly what the Coalition expected. “I think companies will unleash their balance sheets and I think consumers will as well if there is a change of government,” Joe Hockey told business leaders in the lead-up to the election.

Business confidence did indeed jump after the election but then fell back. Consumer confidence jumped and then plunged. The budget sent it deeply into negative territory. It’s now worse than at any time under either Rudd or Gillard.

But things probably aren’t as bad as you think. I know this because unemployment is one of those unusual fields about which the people closest to it often know the least.

I’ll explain with the help of the Melbourne Institute of Applied Economic and Social Research. It has just published an eye-opening study entitled Re-employment Expectations and the Eye of Providence.

The "eye of providence" is the statistical likelihood of being re-employed based on data. "Expectations" are the predictions made by unemployed people themselves.

Sonja Kassenboehmer, of Monash University and Sonja Schatz of Germany’s University of Duisburg-Essen examined the results of a regular survey that asks unemployed Germans to estimate how likely it is that they will score a job in the next two years.

Of those who had been out of work three to five years roughly two-thirds think they are less likely to be re-employed than they actually are, some of them far less likely. Those who believed they had a zero chance of finding work within two years turned out to have a 40 per cent chance. Those who believed they had only a 35 per cent chance turned out to have a 45 per cent chance.

Oddly, the most hopelessly pessimistic underestimaters were people who had been unemployed before. The next most chronic were married people, people who had worked as managers, technicians or clerks or in shops and then people judged as “agreeable” on a psychological questionnaire.

Underestimation matters because people who underestimate their likelihood of being re-employed accept worse job offers. They are more likely to give up searching altogether and if they do keep searching are more likely to  settle for lower-paying jobs, typically taking home $2000 less than people with realistic estimates of their chances of re-employment. And they are less satisfied with those jobs.

Kassenboehmer and Schatz propose a simple fix: "Case workers should ask individuals about their subjective re-employment expectations and inform them of any discrepancies due to underestimation”.

As they put it: “By providing employment agency clients with objective re-employment probabilities, suboptimal behaviour in the labour market such as accepting low-quality jobs or exiting the labour force may be prevented”.

At first they thought it would be complicated. The computer model used by employment agencies would have to take account of things such as personality, gender (women are more likely to underestimate than men), family circumstances, occupation and so on.

Then they examined what they could leave out and still produce a better forecast than unemployed people themselves. It was nearly everything. They usually only needed one  - the number of years an unemployed person had been in and out of work.

It’s information the agencies already have. Feeding it into a computer program could feed their clients reality and lift their spirits as well.

The government mightn’t be up for it. It might be keener on requiring then to apply for 40 jobs a month.

But even if it's not there's something job service providers can deliver - five short words: It’s better than you think.

Peter Martin is economics editor of The Age

Twitter: @1petermartin