SHORTLY after six o'clock on the morning of December 7, 1992, as James Mullins prepared to fire up an illicit Viceroy Ultra-Light in a mine called Southmountain No. 3, in Wise County, Virginia, he probably did not realize that he was about to figure in a debate about the true cost of coal to American society. Almost certainly Mullins did not realize that the operators of Southmountain No. 3 would in the future be charged with thirty-three violations of mine-safety rules, or that by the time he lit up, eight of those violations had allowed a dangerous concentration of odorless methane to accumulate in the shaft where he was working, some 1,000 feet under the surface. The spark from his lighter touched off the methane, which in turn touched off the coal dust floating in the air near the mine's machinery. The force of the explosion leveled a building just outside the mine at the end of a mile-long tunnel. And it killed eight miners, including Mullins. One of them had his skull fractured; the others smothered in a cloud of carbon monoxide. [ILLUSTRATION OMITTED] The national media covered the disaster for a few days, until the bodies were recovered and it was clear that there would be no dramatic rescue story from Southmountain No. 3. No one pointed out to viewers or readers the link between the miners' deaths and the television they were watching or the light that illuminated their newspapers. The fact is, though, that today coal provides more than 55 percent of the electricity generated in the United States. In a high-tech society much rests on the shoulders of men like James Mullins, who descend into dark tunnels, crouch under six-foot ceilings, rip coal from the bowels of the earth, and occasionally die. Twenty years ago few people figured that as the new millennium approached, America would be so dependent on a source of energy that was banned in the city of London in 1273 for being injurious to public health. In 1973 the Federal Power Commission predicted that coal's share of U.S. electricity generation would decline from 46 percent to 30 percent by 1990. Then came the Arab oil embargo and Three Mile Island. Oil became too costly, and too foreign, for extensive use in the generation of electricity. Nuclear power, which had been expected to surpass coal as an electricity generator, was frightening. Coal had the virtue of being widely available inside the United States. And at a cost of less than two cents per kilowatt-hour of power, it was cheap. It was, in fact, so much cheaper than renewable resources like wind and solar energy that it all but eliminated them as commercial sources of electricity. [ILLUSTRATION OMITTED] Or at least coal seemed cheap, until environmentalists, state utility regulators, and some economists began to argue that the market price of coal does not reflect a wide range of 'external' costs that society, or some segment of society, will eventually pay. The economists maintain that the coal James Mullins was digging, which cost about $20 a ton at the mouth of Southmountain No. 3, would be much more expensive if these costs were added on, or 'internalized.' TO begin with, there is the cost of lives lost and health damaged among the miners. The number of deaths tends to decline a bit each year, but that is in part because automation keeps reducing the number of miners. Over the past three years one statistic has remained constant: for every five million man-hours worked in the mines, one miner dies. Last year that meant fifty-four deaths. Society has contrived to place severe limits on the amount of money that miners' widows and children can collect for these casualties. In Virginia, as in most coal-producing states, the law makes it all but impossible to sue mine owners--even in cases like that of Southmountain No. 3, where federal investigators established that safety violations contributed to the miners' deaths. Instead the survivors get a workmen's-compensation benefit calculated on the basis of the dead man's salary and paid by insurance the mine owner must carry. For the survivors of Southmountain No. 3 victims this tops out at $434 a week for 500 weeks, or a total of $217,000. Were the survivors of the Southmountain No. 3 miners able to sue the mine owners, a jury would no doubt place a much higher value on the lost lives. According to a Pennsylvania group called Jury Verdict Research, which tracks such things, $217,000 is about the going rate for damages for a broken leg in American lawsuits. The average jury in a wrongful-death suit awards $941,700 to the family of a forty-year-old man making $25,000 a year. Some economists in the external-costs debate would argue that the dead miners were compensated for their deaths by the wages they earned while they were alive, which were higher than wages in less dangerous industries. In Wise County, for instance, workers in furniture factories make an average of about $7.50 an hour; the miners who died in Southmountain No. 3 made $11 or $12. But the argument that miners consciously calculate that an extra three or four dollars an hour is worth risking their lives for assumes careful, long-term thought on the part of men who are prepared to smoke inside a coal mine. It seems more likely that the extra money they earn compensates them for the immediate travail, the darkness and the dirt and the backaches, of work in the mine. The difference between workmen's-compensation benefits and the market value of a wrongful death--roughly three quarters of a million dollars per case--should be counted as an external cost that the mining industry and the consumers of coal slough off on the miners' families. So should the disparity between workmen's compensation and the market value of the more than 9,000 serious injuries coal miners suffer each year. In the past twenty years the system has internalized at least some of the costs of the miners' occupational disease, black lung. In 1969 federal legislation required that mine operators reduce the amount of coal dust in the air inside the mine to three milligrams per cubic meter, a standard lowered to two a few years later. The operators met the goal by improving mine ventilation and by installing new machinery that sprays a mist of water onto the coal seam while a drill bit gouges out the coal. The legislation imposed a tax of $.55 to $1.10 per ton of coal to pay compensation to miners who had already developed black lung, and it required mine owners to carry insurance to compensate miners who would develop it. To a degree, the legislation has succeeded. In 1969 some 40 to 60 percent of the miners examined had black lung. Among miners who began their careers after 1969, the incidence of black lung has dropped to between five and 15 percent. Still, the compensation to victims of black lung is stingy: a disabled miner with dependent children gets about $200 a week. A CALCULATION of the full cost of coal needs to take into account damage to the environment around the mine. Until the mid-1960s coal-mine owners were from an environmental standpoint barely regulated. The waste from their mines drained into local streams. When strip-mining came along, they began tearing the tops off green Appalachian ridges, gouging out as much coal as they could easily get to and leaving great unhealed brown gashes. When a young biologist named Phil Shelton arrived in Wise to teach at Clinch Valley College, in 1970, he found some of the streams running black with mine wastes. The fish in a nearby reservoir had all died from high acidity. The local residents, taking their cue from the mine operators, seemed to have lost all sense of land stewardship. As with black lung, federal legislation forced mine operators to internalize some of the costs to the environment. Mine operators nowadays have to dig ponds to catch the water running off from their operations, and they must let sediment and heavy metals settle to the bottom rather than pollute the watershed. Strip-mine operators are required to try to restore the original contour of the land, replace the topsoil as best they can, seed it with grass and legumes, and return to plant trees. Seen from a distance, these reclamation efforts look impressive. After a few weeks grass starts to sprout at a reclaimed strip-mine site. After a few years the grass is knee-high, and scrubby trees begin to grow. After fifteen years a properly reclaimed site might be covered with fescue, occasional wild daisies, and spindly pines. But this is not, of course, the vegetation of the original hillside, which might have combined oaks, hickories, and tulip poplars. And when Shelton scoops up a handful of dirt at a fifteen-year-old reclamation site, it is gray and gritty, unlike the black soil, thick with worms and roots, that naturally occurs on an Appalachian ridge. The streams also look better than they did twenty years ago. Bee Branch and the Guest River, the creeks that drain the hillsides around Southmountain No. 3, sparkle in the sun. Their waters are clear and tasteless, and recent studies by biologists from Virginia Polytechnic Institute found that they have no acidity problem. But the streams support almost no fish. Shelton speculates that during spring rains enough settling ponds briefly overflow to fill the creeks temporarily with sediment, stuffing the gills of any fish and killing them off. The records of the local office of the Virginia Department of Mines, Minerals and Energy support his hypothesis. The office receives about ninety complaints a year about mines that have drainage problems or that illegally discharge effluent. Generally, if the mine operator fixes the problem, no further action is taken. By then, of course, the damage may be done. But how to assess that damage? Shelton has heard from his father that before the First World War--that is, before extensive mining--the mountain streams in southwestern Virginia ran so thick with bass that his Boy Scout troop, on a camping expedition, could poke sticks into a pool and gig them. But there are no records of a commercial fishing industry in the local histories Shelton has read and no lost revenues that an economist could confidently charge to coal mining. 'The cost of lost fish?' he asks. 'How do you count that sort of thing?' As a ton of coal leaves Southmountain No. 3, the costs associated with it mount--and most of these costs do appear in the selling price. Trucks take the coal to a preparation plant that crushes it and washes away some of the sulfur and dirt. Then the owner of the preparation plant mixes it with coal from other mines, loads it into trains, and ships it to any of more than half a dozen utilities on the East Coast. The cost of preparation and shipment will more than double the internalized cost of the coal, to about $46, by the time it reaches its customers. Transport adds external costs as well. No one keeps statistics on the number of people hit by coal trains, but coal fills 27 percent of U.S. railroad freight cars and in 1991 there were 423 deaths involving freight trains at public grade crossings. Perhaps 200 trespassers on the tracks died when struck by freight trains. So, roughly gauged, coal's external costs might include about 170 transport deaths a year. TAKEN together, though, the external costs associated with mining and transporting coal would probably add no more than a dollar or two to the price per ton. It is after the coal arrives at a typical generating plant, like Potomac Electric Power Company's Chalk Point Generating Station, in Maryland, that the external costs rise dramatically. Chalk Point is a vast complex of buildings and conveyors, cooling towers and smokestacks, holding ponds and railroad tracks, designed to pulverize coal and blow it into a huge boiler, where it bums. The heat changes water from the Patuxent River into steam that turns generators that, among other things, help keep the lights burning in the White House. Since opening Chalk Point, in 1964, Pepco has, under prodding from federal legislation, made two major changes that internalized some of the coal's environmental costs. It added electrostatic precipitators to the boiler's exhaust system. These capture more than 99 percent of the fly ash before it can escape, and Pepco buries the ash in a landfill. The utility also retired the original smokestacks and replaced them with higher stacks that disperse the exhaust over a wide area and reduce the risk to the health of those living in the immediate vicinity. But the high stacks do not reduce Chalk Point's annual emissions: 50,000 tons of sulfur dioxide, 25,000 tons of nitrogen oxides, and five million tons of carbon dioxide. Debate rages over the assignment of costs to these gases. Sulfur dioxide and nitrogen oxides are important precursors of acid rain, and electricity generation accounts for about a third of the nitrogen oxides and two thirds of the sulfur dioxide America emits each year. This means that coal causes about half the nation's acid rain. But whose coal? Pepco maintains that most of Chalk Point's sulfur dioxide blows over the ocean. Specific sources of sulfur dioxide cannot be tied to specific acidic raindrops. Even if that were possible, it would be difficult to assess damages. The National Acid Precipitation Assessment Program, in an exhaustive survey completed in 1991, looked for damage ranging from crop failures to the corrosion of Civil War monuments. It found relatively little for which the blame could conclusively be placed on acid rain. Some lakes and streams at higher elevations in the East are too acidic to support fish. A few species of trees are suffering, mainly at higher altitudes. But other forest trees seem unaffected, and the fish problem, the study found, can largely be alleviated by periodically spreading lime in the affected lakes. Nitrogen oxides also contribute to the formation of ozone, a known irritant to the respiratory system. But all the diseases related to ozone have many potential causes. No one has devised a precise way to isolate an ozone precursor and attribute to it a definite number of asthma attacks or coughing spells. [ILLUSTRATION OMITTED] The cost of carbon dioxide is still more difficult to pinpoint. It is a greenhouse gas, and there is 25 percent more of it in the atmosphere than there was before the use of coal as a boiler fuel became widespread. But will that 25 percent significantly warm the global climate? And if it does, what damage will that cause? There is no consensus on those questions. Nevertheless, beginning in the late 1980s the utilities commissions of several states decided that they wanted to take coal's external costs into account in their decisions about future sources of power. They used a cost that was calculable: the cost to reduce or eliminate a given emission. In a perfectly rational economic universe the damages caused by a particular source of pollution would always be calculable, and the cost of controlling or eliminating the pollutant would always be less than or equal to the cost of the damages. A utility would not, for instance, spend millions of dollars to overhaul a smokestack that caused paint to peel on a neighboring house. It would simply pay to repaint the house. In the real world, where it is impossible to calculate damages precisely, governments are quite capable of ordering control measures that cost more than the damages they are designed to stop. But for the moment states are using 'control costs' to measure the external cost of coal. California, a state with severe and persistent air-pollution problems, has led the way. New York, Massachusetts, and at least fourteen other states have also tried to assess external costs. None has tried to add external costs directly to the consumer price. But they have directed the utilities in their areas to consider external costs, at least on paper, as they choose among possible new sources of power and present costs per kilowatt-hour to state review boards. Southern California Edison, for example, has been told by the California Public Utilities Commission to assume that in the unlikely event it wished to build a coal-fired plant in the area it serves, each ton of sulfur dioxide emitted would be assessed at $23,490, each ton of nitrogen oxides at $31,448, and each ton of carbon dioxide at $33. Assessments are based on how far within the state's air-quality standards a region is; since southern California is already violating those standards, virtually all emissions from a new plant would have to be significantly reduced, and the costs would be very high. In Massachusetts the external-cost assessments are lower: $1,700 per ton of sulfur dioxide, $7,200 per ton of nitrogen oxides, and $24 per ton of carbon dioxide. Such assessments dramatically diminish coal's attractiveness as a generating fuel when utility companies weigh how to meet new demands for power. If Massachusetts external-cost assessments are used in an example, the ton of coal that costs $20 at Southmountain No. 3 and $46 delivered to the Chalk Point station will be priced instead at roughly $200. Robin Walther, a senior economist at Southern California Edison, calculates that if the state utilities commission's highest external-cost estimates were added in, coal would cost seventeen or eighteen cents per kilowatt-hour rather than around 1.8 cents. Not surprisingly, Southern California Edison has taken the position that actual damages, rather than the price of controls, should be the basis for calculating external costs. Other utilities back an alternative to external-cost assessment which was embodied in the 1990 Clean Air Act limitations on sulfur. Under this system utilities and factories will need federal permits for sulfur emissions that exceed a certain threshold. They can buy and sell the permits, and the market value of a permit to emit a ton of sulfur is deemed by many economists to be a substitute for an external-cost assessment, on the theory that utilities will bid up the price for a permit to emit a ton of sulfur to a level just below the cost of eliminating that emission. Partly because sulfur emissions below the threshold level are free, the permit system promises to cost utilities far less than the external costs calculated by California and Massachusetts. But under any external-cost scenario, coal's price advantage erodes. Natural gas becomes a more cost-efficient generating fuel, since it emits less sulfur. More important, renewable resources like geothermal energy, wind, and solar energy, at five to twelve cents per kilowatt-hour, suddenly become competitive, because they carry few or none of the external costs associated with coal. FOR the moment the impact of these calculations remains theoretical. Demand for power has not grown much in the past few years, and utilities have met their needs largely by encouraging reductions in demand with programs that pay consumers to buy efficient light bulbs or sometimes forgo air-conditioning. Over time, though, the situation will change, as generating stations come due for replacement. A recent study by The AES Corporation, a consulting firm in Arlington, Virginia, predicted, using Massachusetts's external-cost estimates, that renewable resources could by the year 2010 account for as much as 50 percent of the country's new generation capacity, with coal dropping to 10 percent. Much will depend on whether the federal government imposes on the states some kind of uniform requirement for external-cost calculations. If it does not, individual state utilities commissions will have to ponder the economic and political consequences of forcing their utilities to use more-expensive forms of power while neighboring states continue to rely on coal. The Reagan and Bush Administrations spent several billion dollars trying to develop new technologies to burn coal more cleanly. But a 1991 General Accounting Office study of that effort found major components over budget, behind schedule, and a long way from having an appreciable impact. The Clinton Administration's proposal for a Btu tax suggested a recognition of the principle that fossil fuels, particularly coal, have external costs. The rapid demise of the proposal suggests how difficult it will be to persuade the public to pay them wittingly.