Tuesday, September 15, 2015

Eleventy. The arithmetic mistake that made them think ChAFTA would boost jobs

Suddenly, it's all about jobs. Hundreds of thousands of jobs, according to the Trade Minister Andrew Robb. "Literally".

And exports. So unimportant have imports apparently become in the China-Australia free trade agreement that it's now been christened the "export agreement" by enthusiastic (or anxious) government MPs.

Robb and his colleagues used the new phrase an impressive 14 times in question time last Thursday. They referred to "jobs" an astounding 59 times - almost once every minute.

In politics you need to keep repeating something until you vomit, one of its practitioners once told me. Only then does it begin to sink in. So Robb and his colleagues would want to be sure of the facts they were repeating ad nauseam, wouldn't they? They would want to be sure the agreement really will create "hundreds of thousands of jobs".

It won't. It will create only a few thousand, according to government's own modelling, conducted by the Canberra-based Centre for International Economics. But you wouldn't know it from the way the modelling has been mangled and butchered by the government.

It began ahead of its release. The Trade Minister sent journalists "highlights" of modelling, but not the modelling itself. Added to an otherwise faithful reproduction of its executive summary were these words: "The modelling shows that between 2016 and 2035 there will be 178,000 additional jobs as a result of the free trade agreements; an average of almost 9000 extra jobs per year."

Which is odd, because nowhere in the report itself is there a mention of 178,000 additional jobs. The report says that by 2035 the agreements will have produced a total of 5434 additional jobs, a long way short of 178,000 ...

And that figure is a grand total, applying to all three North Asia free trade agreements; those covering Japan and Korea as well as China. Rather than "hundreds of thousands", the total is expected to not quite reach 6000.

By way of comparison, each month the employment total moves up or down by around 10,000 or 20,000 or 30,000; 6000 after 20 years will be something less than a rounding error.

By 2035 Australia's workforce will exceed 15 million. An extra 6000 will be less than one-half of one-tenth of 1 per cent.

But the agreement is all about jobs.

How could the government get its own report so ridiculously wrong? It added up each of the gains reported to employment for each of the years between now and 2035. In 2016 the number is 7925; in 2017 it's 11,119; and so on. By 2020 it peaks at 14,566, and then falls, so that by 2035 it's only 5434 extra.

But these aren't extra jobs per year, they are totals achieved by that year, as the appendix to the report makes clear.

Each of those numbers is already a total. By totalling the totals, Robb and his ministers have been double, triple, quadruple counting, right up to the power of 20.

And there's no doubt that's what they've done. Here's employment minister Eric Abetz in Parliament last Tuesday: "These trade deals will create almost 9000 jobs per year and create 178,000 jobs by the time all the agreements come into full force in 2035. This is visionary, this is wealth and job creating. This is providing a real, positive future for job seekers."

It's nothing of the sort according to the government's own modelling, and the reason it's not is that jobs can't be really be created by trade deals, not in general equilibrium modelling anyway.

The models assume that employment always moves back to its long-run non-inflationary equilibrium, regardless of the new projects and new agreements that are flung at it. Too much of a boost to jobs from a new agreement, and the Reserve Bank will push up interest rates to wind back inflation. Too little a boost and the bank will cut rates to boost jobs.

Trade agreements have little to do with jobs, whatever our ministers (and their soon-to-be-launched advertising campaign) say. And this one has relatively little to do with exports, despite their poll-driven rechristening of it as the "China-Australia export agreement".

Treasurer Joe Hockey told Parliament last week that China spends twice as much on Australian exports as Australia spent on Chinese imports. "It works in our favour by two to one," he said. "The trade agreement that we have with China today is going to make that even better."

No it won't. The modelling says the three agreements taken together will boost total Australian exports by 0.5 per cent. They'll boost imports 2.5 per cent. They are more like "import agreements" than export agreements. They will make Australia better off by encouraging more imports, but on the government's own figures they will send the trade balance backwards.

The modelling finds some industries will do well from the deal - dairy, meat and sugar among them. Others, mainly manufactures, will go backwards. It's worth doing if labour market protections can be built into proposals to import Chinese workers. But there's far less to it than we are being told. Twenty years on, we'll be glad it's in place, but we'll scarcely notice.

In The Age and Sydney Morning Herald