"This is so important,” Sen. Maria Cantwell, D-Wash., said about reauthorizing the Export-Import Bank, “because 85,000 Washington state jobs hang in the balance. That is how many jobs have been supported here in the state by the Export-Import Bank.

How does the senator know how many jobs a few taxpayer-backed loan guarantees from Ex-Im “supported”? She doesn’t. But Ex-Im defenders constantly cite specific jobs figures like this.

These numbers — and all the arguments that Ex-Im creates jobs — are deeply flawed and should convince nobody.

In brief: Ex-Im uses a nearly worthless methodology to count “jobs supported.” Ex-Im subsidies often hurt U.S. employers by subsidizing their competitors. And at best, Ex-Im moves jobs around the economy, boosting employment in some parts of the economy and costing jobs in others.

There is no good way to count jobs “created” or “supported” by Ex-Im’s subsidies — but Ex-Im and its defenders tout such numbers nonetheless. “In 2013 more than 205,000 jobs were created because of financing provided by the Export-Import Bank,” Rep. Denny Heck, D-Wash., said this year in a press release.

The Government Accountability Office in 2013 explained that these numbers are weak approximations using ancient data, crude approximations and nothing approaching cost-benefit analysis.

Ex-Im doesn’t measure how many jobs its subsidy created or many people worked on a subsidized project. Instead the agency does some dubious math. Ex-Im begins by counting the value of exports that it covered with a loan, guarantee or insurance product.

Then it designates which industry that export falls under — aircraft, agricultural machinery, etc. Then it refers to a table created by the Bureau of Labor Statistics, which gives the ratio in any industry between sales and jobs.

For instance, for every million dollars in revenue the pharmaceutical manufacturing industry makes, it employs about 3.51 people. Other industries have higher ratios.

Ex-Im’s claim of 205,000 jobs supported is the result of breaking its export subsidies down into industries and multiplying the industry dollar amounts by the ratio provided by BLS.

The first problem with this counting method: A “job” isn’t necessarily a job. “The employment data are a count of jobs that treats full-time, part-time, and seasonal jobs equally,” the GAO reported.

Second, you can’t take averages from an entire multi-billion-dollar industry and assume they apply for an individual million-dollar export. As GAO puts it: “The data assume average industry relationships, but Ex-Im’s clients could be different from the typical firm in the same industry.”

The marginal Boeing 777 order doesn't necessarily create as many jobs as the first.

Also, that BLS's data is unreliable — it creates its tables using 2002 data.

Most importantly, when lawmakers like Heck say Ex-Im “created” hundreds of thousands of jobs, they are presuming — without evidence — that the subsidized sales wouldn’t have happened absent the subsidy. As GAO put it, Ex-Im’s job counting “approach cannot answer the question of what would have happened without Ex-Im financing.”

It’s absurd for Ex-Im to take credit for every export it subsidizes. The agency, for instance, claims it will support 100 jobs with a $15 million loan guarantee to subsidize Caterpillar’s export of power generating equipment.

But the buyer, Energyst Cat Rental, is a Dutch company created by Caterpillar and other Cat dealers for the purpose of renting out Caterpillar power equipment. Caterpillar owns part of Energyst. It’s not as if Caterpillar was at risk of losing this sale.

Ex-Im subsidies sometimes help a company win a sale it would have lost, but often the subsidies just pad the profit margin of the exporter, or — more likely — the foreign company and the bank that gets the loan guarantee.

Some Ex-Im subsidies directly hurt U.S. employers — such as subsidies to foreign steel mills or semiconductor plants hurting U.S. steel mills and semiconductor makers.

The biggest reason to doubt Ex-Im’s jobs numbers is that government subsidies generally don’t create jobs or profits — they move money around the economy. As the American Action Forum (generally an Ex-Im supporter) put it in 2011:

“Ex-Im’s financing may help create jobs in specific industries. However, for the economy as a whole export financing merely redistributes jobs across the economy rather than create more overall jobs.”

Ex-Im defenders will continue trumpeting these jobs numbers. Nobody should believe them, though.

Timothy P. Carney, The Washington Examiner's senior political columnist, can be contacted at tcarney@washingtonexaminer.com. His column appears Sunday and Wednesday on washingtonexaminer.com.