Greater Baton Rouge Business Report: Go forth. Multiply. Economic developers love to count jobs and

In 1997, the consulting firm KPMG conducted an economic impact study for the expansion of Baton Rouge's Centroplex. It predicted $22 million in annual direct spending by the facility, visitors and exhibitors.

Adding spinoff spending, though, the overall impact swelled to $59 million. Ripple effects like those are what make economic development projects so enticing to governments and the public. Small pebble, big waves.

[ILLUSTRATION OMITTED]

But how are the waves calculated? It is a science that lends itself readily to sleight-of-hand.

Greg Albrecht, chief economist for the state Legislative Fiscal Office, says a properly done economic impact analysis uses a "bill of goods" approach, in which total projected economic output is broken down into categories such as materials, telecom and transportation.

Then each direct impact is plugged into a matrix of geographically specific multipliers. The standard multipliers come from RIMS-II--Regional Input-Output Modeling System--which is run by the U.S. Department of Commerce.

Nothing sells an economic development initiative like the promise of jobs, so employment multipliers are a favorite of project advocates. KPMG's 1997 Centroplex study reported the expansion would directly create some 50 full-time jobs. Including spin-off impacts, total jobs rose 14-fold to 770.

The consultant presented this as a causal relationship: additional dollars spent yield additional jobs created.

But that's not what the numbers mean, Albrecht says.

"We pretend like it's a marginal relationship, but these are really ratios at a snapshot in time," he explains. "Multipliers assume that if I get 10% more income, I'll spend it exactly as I am now. But when a firm gets 10% more revenue, it does not just hire 10% more employees."

Taking such ratios as literal projections of marginal increase can lead to head-scratching results. Last year, CG Railway announced plans to locate a cargo facility in New Orleans. The company said it would employ 10 people, but the state economic development department's figures claimed it would create 600 jobs, and the regional group Greater New Orleans Inc. upped the ante to 1,100 jobs.

The project's boosters got their figures by calculating revenues (the discrepancy in their projections stems from different rail-traffic assumptions). The RIMS-II model says that CG Railway's estimated $60 million in annual revenues would correlate to 600-plus jobs a year.

But in the real world, firms change and adjust. If they have extra capacity or can increase productivity, the marginal effects of CG Railway's move will be far, far less.

"The projections have to pass the laugh test," is how emeritus LSU economist Loren Scott puts it. His consulting firm regularly does impact studies.

As another example, Scott points to the American Sugar Alliance's claim that 29,700 Louisiana jobs rely on the state's sugar industry, a figure being used to argue against the proposed Central American Free Trade Agreement. He figures that would represent about 14 spin-off jobs for every one direct job.

"That's absurd," he says.

Overall economic impact numbers can also be misleading, because advocates have great latitude to choose rosy projections.

They usually choose the output multiplier, also known as business sales. This measures gross spending at every step of the production process.

Say the Saints purchase a practice sled for $10,000. The output multiplier counts that sum, as well as the $8,000 that the sled manufacturer paid to acquire the piping and the padding to put the sled together, plus the $6,000 those various suppliers paid to get the raw materials. A $10,000 purchase thus yields a final output of $24,000--a 2.25 multiplier.

At the state level, RIMS-II output multipliers typically range from about 1.5 to 2.5.

The problem, Albrecht says, is the output multiplier double-counts revenues (or triple-counts, or quadruple, or more) by never subtracting costs. With a full accounting of each sled transaction, things would look rather different.

The Saints' sled supplier paid $6,000 to sell his $8,000 sled, so he only netted $2,000. The pad and pipe suppliers netted the same. The $6,000 in raw materials brings the net effect on the economy to the $10,000 that was actually spent, for a multiplier of 1. And whatever part of the supply chain takes place out of state will pull Louisiana's multiplier down.

This is the earnings multiplier, and it tallies how much of a business' demand ends up in the pockets of local people, either as wages or profits. RIMS-II earnings multipliers for Louisiana range between about 0.3 and 0.7.

"The earnings effect reflects the real impact on the economy," says Scott.

Scott says his clients usually give preference to the more meaningful earnings figure.

Yet when a recent study by Scott for the Casino Association of Louisiana made it into the media, it was simply reported that Baton Rouge casinos contribute $149 million in area business sales and $44 million in household earnings. The numbers were offered without explanation, leaving the casual reader free to assume--wrongly--that they could be added together. Scott says the general public is savvy enough to know what the numbers mean.

Even using the more conservative earnings multiplier fails to recognize that new business activities also have costs, Albrecht contends. If a new employer pushes up wages or absorbs market share, someone takes a hit.

"Each of those downsides also has its own series of negative multiplier effects," he says. But economic impact analysis typically considers only positive outputs.

"It's not that they're dishonest," Albrecht says of the consultants, "but the client wants the biggest number possible, and you're in competition with other consultants."

HAL COHEN covers real estate and legal issues. Reach him at hcohen@businessreport.com.