The Texas Association of Business—the state’s largest business advocacy organization—cites economic studies of dubious quality in projecting a loss of $964 million to $8.5 billion of gross state product if Texas were to strengthen its Religious Freedom Restoration Act in its current legislative session.

In reality, the economic impact of changes to this law is almost certainly too small to be identified in macroeconomic data.

The Texas Association of Business report should not be taken seriously in the ongoing public debate due to its confusion of one-time, annual, and multi-year impacts of supposedly similar policies in other states, along with various other problems.

Here are some of the worst problems with the report:

• The estimated economic impact based on Arizona’s policies can be traced back to a Center for American Progress report based on meeting cancellations in only the first four months following that state’s 2010 immigration enforcement law, extended to cover a period of three years at a rate roughly three times what was actually reported.
• The Texas Association of Business report extrapolates from pure speculation that New Orleans would lose a whopping 85 percent of its major convention business in the wake of former Louisiana Gov. Bobby Jindal’s 2015 executive order on religious freedom, when the city in fact set a record for visitor spending in that year.
• The report claims that Indiana suffered “$1.5 billion of short-term losses alone” as a result of its Religious Freedom Restoration Act, when its source actually said “net present value.” This is a technical term that represents the amount of money that would need to be set aside today to cover all current and future losses associated with the policy, not just the “short-term losses” that the report claims.
• The report refers to a Forbes article that lists a number of events, investments, and hiring decisions that were reportedly cancelled due to North Carolina’s House Bill 2 “bathroom bill,” but aggregates a mixture of one-time and recurring impacts into a $630 million figure that does not represent the economic impact over a well-defined time period.
• In the case of Florida, the report relies on a study that erroneously applies research on lifetime rates of anti-LGBT discrimination to calculate an annual economic impact, while also implicitly assuming in their calculations that non-discrimination policies have no effect whatsoever on the likelihood that an LGBT person will choose to live or work in a particular area.
• Finally, the report cites a non-scientific poll of 212 subscribers to a magazine with a circulation of 50,000, claiming that two-thirds of meeting planners will consider LGBT-related laws in their site selection decisions, ignoring the fact that the questions were worded to attract a sympathetic response.

A small number of vocal sports organizations, entertainers, and businesses on the cultural left have allowed what the Texas Association of Business calls “perceived anti-LGBT legislation” to influence their site location decisions. But these effects are negligible in the aggregate.

It is unclear why Texas or any other state should allow mere perceptions—not reality—to establish its social policy, particularly on a matter as fundamentally important as protecting the constitutional right to freedom of religion.