Public pensions are complicated beasts. They represent the aggregation of promises made to public employees—both current and former—to pay them benefits from their retirement until their demise. They sum to a growing—albeit somewhat imprecise—stream of payments that presently extends to close to the end of the current century. The predilection of politicians to make promises that won’t come due until long after they’ve left office has resulted in many states with promised benefits far in excess of the money set aside to meet them.

Many state pension authorities managing underfunded pensions—often nudged by their legislatures—have aggressively invested their pension assets in an attempt to achieve supra-normal returns and reduce the shortfall, which would be infinitely preferable to them than increasing taxes, reducing benefits, or trimming other government spending.

But such a strategy amounts to a sucker’s game. On average, putting money in hedge funds—or practically any actively-managed stock fund for that matter—has returned less, after fees, than investing in a diversified stock/​bond portfolio and does little to limit the downside risk should a sustained bear market develop, let alone another fiscal crisis.

For this reason, more prudent states have made an earnest attempt to reduce their management costs and risk by embracing passive investing, whereby the bonds and stocks owned consist of a broad-based index of the entire market. In essence, portfolio managers have no real decisions to make when they invest in a market index, so the management costs are minimal.

While a strategy that eschews actively selecting a portfolio may offer the highest long-term returns with the least risk, some object to such a strategy for a government pension. Holding stock of every single member of the S&P 500—the most common index used in passively-managed index funds—entails investing in companies and industries that some find objectionable. Activists have objected to companies that they perceive as being insufficiently welcoming to gay and lesbian workers (Cracker Barrel), or who pay a portion of their workforce below what they deem a livable wage (Walmart), or whose senior management has committed personal peccadillos (Abercrombie & Fitch), or treated women employees poorly (Tesla), or breached one of a number of other myriad precepts put forth by such groups. These days, some non-profits perceive any active participation with the current Trump Administration as being objectionable.

Perhaps the most objectionable sorts of investments for liberal activists these days consist of companies that extract and sell fossil fuels. Many perceive climate change, abetted by the burning of fossil fuels, as an existential threat to the planet and fear the actions taken by the governments across the world—and especially the United States—to be insufficient to effect a change.

The frustration that many feel about climate change inaction by the federal government has led them to pursue change at the state and local level. This political pressure has led to 34 states creating their own climate action plan, and 29 states now mandate that utilities increase the proportion of renewable energy they purchase. California has gone even further and has imposed strict new limits on emissions from new automobiles.

Another way that activists have pushed state and local governments to fight climate change has been to push for government pensions to eschew investments in companies that extract fossil fuels. Such a policy can be complicated since most index funds merely hold a basket of the stocks of companies in the exchange, these become out of bounds. Such a policy asks an important question: Can a pension fund avoid fossil fuel investments or other objectionable investments and still earn a comparable rate of return with minimal risk, as in an ordinary index fund?

The answer is no.

These days there are are a number of variants of “socially responsible” funds to help people avoid investing in companies whose business model, industry, or ethical practices they find disagreeable. Some of these funds are also passively managed and come with relatively low fees. For instance, in 2015 the S&P 500 created a fossil-fuel-free index, and investors can effectively invest in it by purchasing a “SPideR” ETF. Vanguard—considered to be the leader in passively-managed funds—offers the Vanguard FTSE Social Index Fund.

Some environmental activists have observed that broad-based, environmentally responsible stock funds that excluded shares of companies that extract fossil fuels outperformed the broader stock market index over the last three years. For instance, Vanguard’s Social Index Fund outperformed its broad market index both in the past 12 months and the past three years. Vanguard’s U.S. fossil-free index, created last year, has also outperformed the broader market since its inception.

These outcomes have led some to suggest that if state pension plans had invested in such vehicles the last few years they could have both higher returns as well as the moral high ground—a win-win situation.

For instance, in 2016 the research firm Corporate Knights concluded that if the New York State Common Retirement Fund, the third largest state fund in the country, had moved away from investing in fossil fuels it would have actually boosted its returns the previous three years by $5.3 billion. It presented these results as an example of one way that states could reduce pension funding shortfalls while helping the environment.

Are public pension managers doing a disservice to pensioners and taxpayers by failing to adapt to a changing economy by investing in companies that are polluting the planet?

In a word, no. Investing in an environmentally or ethically conscious way will invariably impose some sort of cost on an investor. While an individual investor may be willing to receive a lower return to assuage his conscience, asking a state’s current and future retirees—or the taxpayers who support them—to do the same in a world of underfunded pensions is a dangerous precept. And to pretend doing such a thing is costless is misleading.

Narrowing the scope of the stock held in a portfolio necessarily increases its volatility. Holding fewer stocks and industries results in a market basket that is more susceptible to market fluctuations. It is possible that the narrower portfolio works in favor of the investor in certain circumstances, but modern portfolio theory suggests that the probabilities are that it will, in fact, underperform the broader market index.

The rationale against such arithmetic is that such a fund excludes companies that produce an asset that is in decline, leaving it with what ought to be, at first blush, a superior-performing mix of companies. However, there’s no reason to believe this is true. For starters, the fortunes of an individual company are not wholly tied up in the fortunes of the larger industry.

For instance, the tech sector has grown exponentially in the 21st century, and companies such as Facebook, Google, and Apple have made hundreds of billions of dollars for their investors. However, investors in Microsoft, which had the highest market cap in the S&P 500 in 1999, have lost money by holding onto its stock.

In the U.S. smoking has been on a long decline both in the U.S. and elsewhere for the last twenty years, and there are only about half as many smokers in the country today as there were at the turn of the century. A similar trend has occurred in most developed countries.

A world where the popularity of its main product is steadily declining would presumably spell disaster for any company, but that has not been the case at all for Altria, the maker of the immensely popular Marlboro cigarettes. Its profits have grown throughout the 21st century, as has its stock price.

In 2000, The California Public Employee Retirement System (Calpers), the largest public pension fund in America divested its holdings in the tobacco industry, citing moral concerns about supporting a product known to be deadly. A number of other pensions followed suit. With mounting medical evidence of the dangers of tobacco and public consumption declining, it seemed like a financially sound decision as well.

As a result Calpers have missed out on as much as $3 billion in gains.

Betting against fossil fuels is no less counterintuitive. ExxonMobil, by far the largest fossil fuel company in the U.S., has a stock price that moves closely with the price of oil, so any success in reducing demand for oil would at first blush harm their bottom line and stock price. However, it has also made a strong play into natural gas—which emits much less carbon dioxide when burned—and it has been increasing its investments in algae biofuels of late. Thus, a political environment less friendly to carbon emissions need not harm its stock price.

It also has a reputation of being an extremely well-managed firm that manages to keep its costs under control under all circumstances, a feat that has earned it the admiration of its industry rivals.

What’s more, the notion that the fortunes—and stock price—of ExxonMobil and other firms in its industry will collectively fall if and when pro-carbon policies and technological changes take off assumes a modicum of market myopia. If there were a collective perspective that carbon taxes are likely in the next five to ten years, the market would have already incorporated such information into its stock price. Given that its stock price is down over 20 percent since its peak in early 2014 suggests that this may indeed have occurred to some extent—which means that Exxonmobil’s fortunes are—as before—subject to its management performance as well as the fortunes of the broader energy markets.

Besides the impossibility of saying with any certainty that a carbon-free portfolio will outperform a broader internet index, the cost of actually constructing such an index results in management fees significantly higher than a simple index fund that attempts to mimic the behavior of the broader market. A market index can be constructed via a simple computer formula, and management fees tend to be quite low. The Vanguard 500 Index Fund, for instance, has an expense ratio of just .14 percent, compared to the average actively managed fund ratio of 1.25 percent.

However, a fund governed by certain moral or environmental principles requires a constant examination of the companies in the index and their behavior, which unavoidably results in a higher expense ratio. The Vanguard FTSE Social Index Fund has a management fee of .22 percent—lower than nearly all actively- managed funds but still 50 percent higher than Vanguard’s 500 index fund. The S&P 500 Fossil Free SPideR has an expense ratio of .4 percent—thrice that of the regular Vanguard Index Fund.

In its 2015 report on the problems of the Fiduciary rule, the Obama White House embraced the use of index funds for long-term investors because of the ultimate advantage of low expense ratios.

Higher management fees ultimately translate into lower returns and the performance of the Vanguard Social Index bears that out. While it did outperform the broader market index from 2012–2015, a period that coincided with a steep decline in oil and coal prices, the five and ten year comparisons show the broader index with higher returns, consistent with theory.

The idea that a state portfolio manage can improve his investment performance by assuming that the future will entail a substantial reduction in carbon, and that this environment will pull down the fortunes of ExxonMobil and other oil producers is facile. It may very well occur, but to assert that with any degree of certainty borders on hubris.

The fiduciary responsibility of a public pension manager ought to be to pursue the maximum long-term return while minimizing the long-term risk. The fact that a government pension is, in a way, eternal introduces a modicum of complications and freedoms into that equation, but it doesn’t change that goal.

A public pension fund’s extended time horizon can allow the fund manager to pursue relatively illiquid investments that may not have a payoff for decades down the road. The tradeoff for that illiquidity should be, on average, a higher return.

Pension funds—as well as insurance companies and other entities with long-term liabilities—strive to match their long-term liabilities with similarly-lived assets, which serves to further reduce their exposure to risk. This preference for long-lived assets is one reason that the U.S. Treasury has begun considering the sale of fifty-year bonds. However, allowing political exigencies to determine a public pension fund’s portfolio complicates such decisions.

There is one crucial difference between a privately-managed hedge fund and a public pension fund. A private investment firm that performs poorly will ultimately lose clients and money if it fails to perform.

A state-run pension fund has no competition, and the consequences of poor performance are usually less dire for such investment managers. For this reason, the danger of political activism, catering to special interests, and making financial decisions based on rationale that go beyond the best interests of the plan participants is exceedingly problematic. That a government can potentially supplement a financially deficient plan with tax dollars may be true—and something the state of Illinois will soon have to contemplate—but it should not be an invitation to use a public pension to carry out a political agenda in a way that exacerbates volatility and reduces potential returns.