Energy-efficiency investments pay for themselves. This oft-repeated claim—upgrade your home’s insulation, seal the cracks, buy an LED, and you’ll eventually turn a profit—is a shibboleth among environmentalists.

Unchallenged for many years, the energy-efficiency mantra has been called into question in recent months, with enraged environmentalists and a controversial academic turning an obscure tiff into a full-fledged wonk-storm. Yesterday, Eduardo Porter of the New York Times penned a scathing attack on a government study showing the benefits of a federal home-weatherization program.

At the core of the argument is a simple question: Is a lower energy bill the only benefit of home weatherization? Or, when calculating the returns on our energy-efficiency investments, should we also include the improvements to health and productivity that home weatherization brings to low-income families?

Here’s some perspective to help you understand the dispute.

An injection of cash.

Energy-efficiency experts believe they can significantly reduce carbon emissions, make people more comfortable in their homes, and improve air quality—all while turning a profit. Policymakers, however, haven’t been enthusiastic. Maybe it’s because they’re skeptical about offers of a free lunch.

The federal government finally gave energy efficiency a big push in 2009 with the stimulus bill, increasing investment in home weatherization by a factor of seven to around $1 billion. The influx of federal money presented a challenge: Would the old mantra about energy efficiency paying for itself hold true at this scale?

There were reasons to believe it might not—money changes everything. The extra cash allowed technicians to go beyond the most critical upgrades. At some point, they may begin to do work that didn’t pay for itself. In addition, abstruse federal budget rules forced a higher proportion of weatherization money to flow into warm, dry states, where it is less needed. The stimulus bill also required that weatherization agencies employ more expensive workers.

As state agencies insulated homes around the country, the Department of Energy began to collect data that would reveal whether energy efficiency really is a free lunch on a national scale. But it would take years for those results to reach the public.

Enter the provocateur.

While analysts at Oak Ridge National Laboratories crunched the data, economist Michael Greenstone strode onto the scene. Belonging to a more modern breed of economist, Greenstone isn’t content to chin-scratch at conferences over the shape of some obscure curve. Instead, the University of Chicago professor tends to spark controversy.

In 2012, for example, when Hillary Clinton and Julia Roberts were pushing clean stoves to improve indoor air pollution in developing countries, Greenstone challenged them. He published data indicating that recipients quickly abandoned their free stoves, either because they broke down or the family was too accustomed to traditional cooking methods.

In June, Greenstone set his sights on energy efficiency, coauthoring a study that examined a weatherization program for low-income families in Michigan. There were two controversial findings. First, the authors found that the weatherized homes saved less than half as much energy as assessors predicted they would. Second, the authors concluded that the program was not the economic free lunch it was touted to be. The government spent $4,143 per weatherized home, but the average family would eventually save only $2,400 on their energy bills as a result of the upgrades. The headline-making finding was that energy-efficiency programs do not pay for themselves.

The New York Times, Wall Street Journal, and Vox were among many outlets that covered the study, recognizing a provocative hook when they saw one. Debunking is the currency of the Internet, and Greenstone’s study hit the jackpot.

Once you read past the headlines, though, Greenstone’s approach raised some troubling questions. He looked at one program in one state and only considered a single measure of benefit: utility bills. Weatherization programs decrease emergency-room visits, prevent missed school and work days, reduce asthma symptoms, prevent climate change, and offer many other values that Greenstone didn’t attempt to quantify.

He may also have inadvertently biased his sample. Most states, including Michigan, have long waiting lists for weatherization. They look for homes with big energy bills—the ones likely to benefit the most from efficiency upgrades. Greenstone, by contrast, spent $475,000 advertising his study. The authors called it an “encouragement campaign,” but it probably drew in participants who wouldn’t otherwise have been chosen by the program.

The study was also published as a non-peer-reviewed “working paper,” which significantly diminishes its credibility. The timing is strange as well: It was widely known that the government was preparing to publish its own data on energy efficiency when Greenstone pushed his paper out.

Based on these limitations, energy-efficiency advocates assailed Greenstone, much like the clean stove advocates had done three years earlier. That’s what happens when you try to tip someone’s sacred cow. But, methodological quibbles notwithstanding, Greenstone kicked off an important conversation.

The big study drops.

Oak Ridge National Laboratories—a research facility operated by a contractor on behalf of the Department of Energy—released its official findings for the expanded weatherization program last month.

The study dwarfed Greenstone’s research. The 4,500-page report was national and looked at many aspects of weatherization—from energy bills to doctor’s visits to home fires to improved sleep—and incorporated dozens of peer-reviewed studies. Unlike Greenstone’s paper, the research itself underwent an extensive, multilayered review process

Until yesterday, the Oak Ridge study received virtually zero media attention. That’s sad but not surprising—it’s long, lacks a controversial hook, and isn’t being pushed by a charismatic, media-savvy professor. “The Benefits of Energy Efficiency Programs Are Complex and Multifaceted” doesn’t have the mass appeal of “Everything You Think You Know About Energy Efficiency Is Wrong.”

The Oak Ridge study, in one way, concurred with Greenstone: Energy savings don’t always cover the costs of weatherization. The influx of cash from the stimulus bill, and the strings that came with it, pushed the cost-benefit ratio down nationwide.

But—and this is a crucial point—that’s only true if you limit the analysis to energy savings. The Oak Ridge study found benefits in other areas. Smoothing out the data across households, the researchers found that weatherization saved the average family $2,009 by reducing asthma triggers, $154 by limiting carbon monoxide exposure, $831 by preventing fires, and $1,329 by improving sleep and work productivity. Remember that these are benefits to low-income families. The sums are significant.

Greenstone slammed the research in yesterday’s New York Times story, which makes two central criticisms.

First, the article argues that Department of Energy scientists are not in a position to review their own work, likening the Oak Ridge study to government inspectors in India who fell under the influence of the manufacturers who selected and paid for them. The comparison is ludicrous. The Oak Ridge study is an open piece of research. The analysts showed all their work and submitted their research to peer review.

Greenstone, in contrast, published a non-peer reviewed report, then aggressively promoted his conclusions in the media. When climatologist James Hansen did precisely the same thing in July, many outlets refused to cover the study. Andrew Revkin, Porter’s colleague at the Times, wrote at length about the problems with Hansen’s decision to bypass peer review. Apparently lacking Revkin’s scientific nous, Porter accepted Greenstone’s views with far less scrutiny.

The second critique is that Oak Ridge overreached when attempting to monetize the noneconomic benefits of weatherization, like those affecting health and productivity. It’s impossible to address these complaints without descending into truly somnolent detail. Suffice it to say that there’s nothing novel about the Oak Ridge methodology, as Greenstone claims. Survey data is not a perfect source of information, but the surveys employed were externally validated and are used by medical researchers. Greenstone himself has relied on survey data in the past.

The larger question, which Porter fails to raise in his coverage, is whether there is a better way to quantify those hard-to-monetize benefits. Greenstone doesn’t propose one. Instead, in his research, he wrote them off entirely so he could produce a media bombshell. In Greenstone’s analysis, nothing but energy savings is worth anything. Reduced exposure to asthma triggers? Worthless. Improved sleep? Who cares. Prevention of home fires? No big deal. Reduced carbon monoxide exposure? Meh. Climate change? Never heard of it.

We need gadflies like Greenstone to check the government’s work. But let’s keep perspective: Just because something isn’t included in a utility bill, that doesn’t mean it’s not valuable. An economist’s job is to find a way to assess that value. Greenstone never even tried.

The takeaway.

We ought to be more careful about how we discuss the value of energy-efficiency investments. The traditional, sweeping statement about every investment paying for itself is too broad. You can’t just keep stuffing more and more insulation in an attic and expect cash to rain down from the ceiling.

But smart efficiency investments do pay for themselves. Well-targeted energy-efficiency programs pay for themselves. Tightly run energy-efficiency agencies pay for themselves.

At the same time, it’s important to recognize that energy-efficiency programs offer wide-ranging benefits—they aren’t just support measures for low-income families. They reduce strain on the power grid, lower carbon emissions, and improve air quality for everyone. You can’t simply ignore those benefits because they don’t come with dollar signs attached.