As the effects of climate change unfold, its impact on business will grow more severe: Altered rain patterns will affect agriculture, floods will disrupt supply lines, heat waves will prevent employees from working. If markets are to work well, investors need to know about these consequences, and the number of companies that voluntarily disclose their estimates of climate-linked risks has risen markedly the past 15 years. They now account for around 69 percent of market capitalization, and such disclosure could soon be mandatory in many nations.

Of course, information only helps if it's accurate. A comprehensive new study has assessed the plausibility of these corporate disclosures, comparing them to the best scientific and economic estimates of the likely costs of adjusting to climate change. The good news: Every year, more businesses start taking climate risks more seriously. However, current reporting also has serious blind spots that could leave investors uninformed and exposed.

In 2015, Bank of England Governor Mark Carney warned in a speech to Lloyd's of London about the "Tragedy of the Horizon," his term for global risk arising from the inherent disparity between the short-term thinking of financial markets and the long-term nature of climate. Companies concerned with pressing challenges, he suggested, may not be accurately disclosing to investors the risks they face due to climate change.

To test Carney's hunch, Allie Goldstein of the environmental organization Conservation International and colleagues took the voluntary disclosures made in 2016 by more than 1,600 large companies worldwide and compared the risk estimates within to estimates from scientists and economists.

According to their findings, most companies expect climate change will increase their operational costs and reduce or disrupt production capacity due to events such as floods, drought or hurricane damage. And awareness is growing rapidly: The number of companies seeing such risks as either "virtually certain" or "more likely than not" to occur surged from 34 percent in 2011 to 67 percent in 2016. Yet serious complacency persists.

Economists have used so-called integrated assessment models, which combine economic and climate models, to make crude estimates of the likely costs that will be incurred dealing with the physical aspects of climate change. Through 2100, this approach leads to figures varying from around $2 trillion to $20 trillion, or 2 percent to 20 percent of current total financial assets.

Independently, the Intelligence Unit of the Economist estimated a possible loss of 30 percent of the entire stock of manageable assets. In striking contrast, Goldstein and colleagues found, companies estimate their financial risk only in the tens of billions of dollars. That's an alarming 100 times smaller than even the most conservative scientific estimate.

It's important to remember that integrated assessment models have themselves been heavily criticized for being inherently conservative, and they probably underestimate the likely costs by quite a lot. So it seems that firms as a group are not even coming close to providing realistic views. Or they're not faithfully reporting the risks they already see, perhaps believing the truth could turn investors away.

Another problem the researchers found: Businesses seem to be taking an overly narrow approach to risks, ignoring the indirect costs that might arise from disruptions to supply chains or changes in customers' behavior. For example, estimates of how rising temperatures will affect productivity by 2100 forecast a 20 percent drop in global income per capita, which would naturally lead to falling overall demand for goods and services. Even so, fewer than 3 percent of companies on their disclosures considered it plausible that climate change could affect their businesses this way.

It looks like Carney was right about financial markets ignoring climate risks, which raises questions about their ability to steer investments wisely. In seeing long-term risks, markets don't seem nearly as efficient as we might hope. This might, as some management theorists have previously suggested, have to do with the culture of business itself, as mainstream business models generally work on the assumption — clearly dangerous in the context of global climate change — that current economic and social conditions will continue into the indefinite future, regardless of what happens to the Earth's environment.

Buchanan, a physicist, is a Bloomberg View columnist. He is the author of "Forecast: What Physics, Meteorology and the Natural Sciences Can Teach Us About Economics."