Nothing has perplexed economic policymakers more than the historically high unemployment rate that has persisted despite two years of reasonably good economic recovery from the worst recession in the postwar era.

The unemployment rate was 7.3 per cent in March, the lowest it has been since the end of the recession in the spring of 1975 (it peaked at 9 per cent in May 1975), but still above the highest unemployment rate recorded during any of the previous postwar economic downturns.

Yet, since the end of the recession, the number of persons with jobs has climbed markedly. In May 1975, there were 84.5 million people at work (and 8.3 million looking for work but unable to find it). By last month, the number of jobholders had increased to 89.5 million. That is an increase of 5 million jobs in less than two years, a good performance by nearly any standard.

While there has been a marked increase in jobs, there has not been as rapid a decline in joblessness, nor as big a drop in the unemployment rate. When the unemployment rate was 7.3 per cent in March, the number of unemployed was 7.1 million.

Both the Carter and Ford administrations came under heavy attack from organized labor and some legislators for not devising big job-creation programs in the wake of massive unemployment.

The continuing high unemployment has prompted the administration and Congress to enact public service jobs programs, accelerated public works projects and numerous other programs designed to increase the number of persons at work.

Economists at the University of Miami argue in a new study that the unemployment picture is not nearly as serious as depicted in official statistics and that public policy ought not be devised with the notion that the nation's jobless rate is 7.3 per cent.

Economists Kenneth W. Clarkson and Robert E. Meiners, argue instead that the "high measured rates of unemployment can be explained in large part by a new class of individuals who are either largely unemployable or have no need or desire to work, but who, to qualify for various welfare benefits, must officially register for work and therefore are now counted in official unemployment statistics."

They say that the "recent upsurges in the official unemployment statistics are the result of the introduction of these work registration requirements." Most of these work requirements have been instituted since 1970 for such programs as food stamps, aid to families with dependent children and general state welfare assistance.

When unemployment statistics are adjusted to account for the work registration requirements, the unemployment picture corresponds more closely to historical trends, Clarkson and Meiners say.

The unemployment rate averaged 5.6 per cent in 1974. Clarkson and Meiners say that after they adjust for the impact of required work registration for recipients of food stamps and aid to dependent children, the corrected unemployment rate is 3.8 per cent.

In 1975, the published rate averaged 8.5 per cent. The corrected rate is 8.1 per cent. Last year, when the average rate was 7.7 per cent, the adjusted Clarkson-Meiners rate was 5.3 per cent.

The Bureau of Labor Statistics is examining the study but has come to no conclusions about it, according to the commissioner of labor statistics, Julius Shiskin. Shiskin said he expects the staff analysis to be finished in a few days. Members of the BLS staff characterize the size of the adjustments as "heroic."

The Miami economists note that during the only period before 1970 in which more than 56.5 per cent of the civilian population had jobs (1969) -- a level they call high employment -- the unemployment rate was 3.5 per cent. During periods of medium employment (when about 56 per cent of the population is at work,) the unemployment rate has averaged 4.4 per cent, while in periods of low employment, (55.15 to 55.57 per cent of the population at work), measured unemployment averaged 4.9 per cent between the end of World War II and 1970.

But in 1975, when the employment rate was 55.25 per cent, the unemployment rate was well above the historically expected rate of 4.9 per cent, averaging 8.5 per cent. In 1974, when almost 57 per cent of the population had jobs, the unemployment rate was 5.6 per cent.

"Based on historical averages, unemployment in December 1976 should have been closer to 4.5 per cent as opposed to the recorded 7.9 per cent," Clarkson and Meiners say.

Economists have advanced many explanations for the continuing and historically high unemployment rates. Some unemployment can be attributed to the higher levels of unemployment compensation. The labor force has changed, too -- with larger numbers of teenagers and women than in the past. Supposedly these two groups have had historically higher unemployment.

But Clarkson and Meiners argue that, in the last five years, there has been no appreciable increase in the numbers of teenagers in the work force and that, while more and more females are joining the labor force, their employment is rising even faster.

Instead, nearly all the explanation lies with the change in work registration requirements.

"Large numbers of individuals must register for work and as a result be counted in the official unemployment statistics," the economists say in their report. "If these individuals generally prefer not to work or are largely unemployable, the work registration requirements will permanently increase the measured rate of unemployment. This means that the unemployment data collectd since the implementation of the work registration programs are not comparable to the data collected before that time."