THERE ARE BOTH reasons for cheer and grounds for some concern in the Environmental Protection Agencies decision the other day to give five auto makers until 1983, instead of 1981, to cut carbon monoxide (CO) emissions of some engine models from 7 grams per mile to 3.4. The good news is that auto technology is advancing, though unevenly. EPA found that General Motors, Toyota and Volkswagen of American have largely managed to develop the intricate engine controls required by the 1981 standard -- or have not shown that they can't. In contrast, Chrysler got extensions for almost half of its cars and American Motors for about four-fifths. As gauges of the companies' technological prowess, the results are about what one might expect.

From a competitive standpoint, the waiver system has some awkward effects. It puts a certain penalty on better engineering; GM estimates that meeting the stricter standard will cost $30 to $40 more per car (EPA says $11 to $17). Is that offset by the market appeal of a more advanced car? One answer to that may be suggested soon by Ford, which is still testing its 1981-model engines and has not yet decided whether seeking waivers is necessary or desirable.

From a public-health perspective, what difference will the extensions make? Very little, says EPA. The interim standard of 7 grams per mile, which takes effect with 1980-model cars, will cut CO emissions substantially in the next few years. Deferring a further reduction for one-fifth of the 1981 and 1982 cars, the amount exempted so far, will not have very noticeable effects on urban air quality.

That points to a much larger question: Is the final cut in CO emissions necessary at all? The 3.4 grams-per-mile goal goes back to 1970, when Congress -- not knowing what levels of pollution are how harmful -- decided to reduce each major category of auto emissions by 90 percent. But the evidence justifying those goals has not accumulated as neatly as many people expected. Carbon monoxide can obviously kill someone who breathes a lot of it in a closed place. But some experts now believe that small concentrations of CO in the open may have much less serious or lasting effects than small amounts of other emissions, such as nitrogen oxides. At some point, choices may have to be made.

It might seem odd to talk about altering clean-car goals that the auto industry may well be able to meet in the next decade. Yet one must also consider tradeoffs with other goals, such as better fuel economy. Finally, the new technology will work only as well as its users demand. Those systems of computers and catalytic converters will require more careful maintenance than many car owners tend to provide. They will also need unleaded gas. And the problems that developed on that score this summer are one more caution against putting too much faith in engineering as an answer to all ills.