ILW.COM - the immigration portal Immigration Daily

< Back to current issue of Immigration Daily < Back to current issue of Immigrant's Weekly

The Lawyer's Guide To 212(a)(5)(A): Labor Certification From 1952 To PERM

by Gary Endelman

Gary Endelman

Introduction: Where We Came Out

Most Americans will doubtless be surprised to learn that the vast majority of legal immigrants working in this country never had a work visa. They came here through family ties free of any labor market controls; their potential impact on the domestic labor force, not to mention their potential contribution to the U.S. economy, was never considered. Americans have traditionally thought of immigration as social outreach or international social work. Asylum, providing refuge, and family unity are all acceptable because they nurture the poor and the downtrodden. By contrast, employment-based immigration has never aroused the same passion. It has always been regarded more as a necessary evil, an embarrassing relative that one does not talk about in polite society. Many Americans, even most immigration advocates, have always secretly felt that business-related immigration lacks sufficient moral integrity to justify their support.

The basic reason the U.S.´s current immigration laws do not protect U.S. workers is the total absence of labor controls on family-based migration, which accounts for the majority of those who come here each year. Labor certification cannot be changed unless all of the employment-based system is also overhauled. Such a reform necessarily would require a radical restructuring of the family preferences to emphasize the integrity of the nuclear family and eliminate chain migration. While most Americans love their siblings and adult children, they do not live with them. Family preference categories should reflect this. The U.S. should no longer tolerate visa backlogs that divide the families of lawful permanent residents for years. The spouses and unmarried minor children of all such permanent residents should be treated as immediate relatives and come into the U.S. without any numerical restrictions. All other family-based categories should either be abolished, with their numbers being transferred to the employment categories, or such migration should be subject to labor certification. This would not prevent the extended family members from immigrating to the U.S., but it would require a proper system of labor controls as a way to promote the national interest and protect the legitimate interests of the domestic work force.

Most Americans care more about family-based immigration than they do about employment-based immigration. They understand and support family unity and applaud the natural desire to keep the family together. By contrast, both the left and the right in U.S. political thought share a nagging suspicion that more employment-based immigration may not be so good for the U.S. after all. Growing the economy in a way that benefits Wall Street or Silicon Valley makes us uncomfortable. Be on the same side as the multinationals? Enlist under the banner of a free-market economy in the age of scandal and corporate fraud? Hardly the stuff to inspire a great moral crusade. Perhaps the reason that some immigration advocates do not invest in fundamental reform of employment-based immigration the same fervor and commitment that they so readily and repeatedly lavish upon family issues is that, when the dust settles, they neither fully trust nor believe in the capitalist system and the values it represents.

Any attempt at labor certification reform can never succeed unless it starts from the premise that employment-based immigration exists not to help the alien but to enrich the U.S. economy. While the U.S. needs to retain a core commitment to asylum and family unity as an expression of what Lincoln called the "better angels of our nature," policymakers must realize that employment-based immigration is an exercise in enlightened national self-interest. The Department of Labor (DOL) has taken a bum rap over the failures of labor certification. The DOL is not to blame for a program whose goals have never been defined by the nation or its leaders. In truth, labor certification is about where we should expect it to be. It is the perfect institutional embodiment of our national schizophrenia concerning employment-based immigration. All of the clashing assumptions, competing priorities, and contradictory objectives concerning labor certification since 1965 make any effective action by the DOL an occasion for relief and surprise.

Whatever one thinks about labor certification, the protection it has provided for U.S. workers is less than overwhelming. The real threat to the legitimate interests of U.S. workers comes not from the inadequacies of labor certification, real or imagined, but from the chain migration to the U.S. through family ties that is entirely legal but largely unchecked by any effective labor screening. While most critics of employment-based immigration zero in on labor certification, the truth is that the number of "green cards" earned this way is very small compared to the much larger number of family immigrants who are really coming to work. This has not become a major issue until now because family visa categories are protected by a widely shared belief in their ethical legitimacy, a belief that has never been extended to any employment options.

The problem with labor certification, then, is, in reality, a symptom of a lack of clarity concerning employment-based immigration in general-what it is and what we want it to do. Until we achieve a lasting and coherent national consensus on that, genuine progress on labor certification is unlikely. So long as the absolute numbers available for migration through employment were limited, the contradictions within the system could be contained. When the Immigration Act of 1990, Pub. L. No. 101-649, 104 Stat. 4978, tripled the number of employment-based immigrant visas, it was no longer possible to contain these contradictions. Unable to exert any influence over most employment that came in as part of the family stream, the DOL not surprisingly resolved to exert maximum control over the few immigrants who required a labor certification. The system was never designed for the large numbers we have today, however. The 1965 compromise on the labor certification language in the INA, which accompanied the abolition of the national origin quotas, is under increasing strain and the DOL has turned to Program Electronic Review Management (PERM) as the last, best hope to keep the labor certification system alive.

As we wait for PERM to be born, let us take time to look at the history of labor certification in order to understand how we got to where we are. Labor certification is sick and the patient has been ill for quite some time:

In the near future, Congress must turn its attention to...the permanent labor certification program.... The current scheme is neither prompt nor efficient, nor are the outcomes predictable or uniform... Criticized by those in government, organized labor and business, the current labor certification program has no defenders and will be replaced as soon as its critics devise an adequate and acceptable substitute... The time has come, for thoughtful analysis and proposals.1

The delay in processing applicants’ labor certifications is a national concern of the first order. Employers cannot plan, workers worry if they will be able to hold on to their jobs long enough for the labor certification to be decided, and children "age out" by turning 21, thereby losing any entitlement to derivative immigration benefits. The Child Status Protection Act, Pub. L. 107-208, 116 Stat. 927, does not cover these children because their parents’ employers cannot file an immigrant petition without an approved labor certification. The DOL has no official policy to expedite such "age out" cases although individual regional offices may do so on humanitarian grounds. Leading immigration lawyers have urged their clients to petition the DOL to rule on their cases for precisely these reasons.

The backlog in processing the labor certification must be eliminated to allow applicants to become productive members of the economy. An efficient labor certification system will allow applicants to file for permanent residence and allow their spouses to work. This, in turn, will expand the tax base, enhance government revenue, and benefit the U.S. economy. A dysfunctional labor certification system is a drag on the economy; a functional system would be a critical component of a system of large-scale immigration. That is why lawyers, employers, and government officials need to pay attention, and why the DOL had to turn to PERM as a last resort. In explaining the rationale for PERM in May 2002, the DOL said as much:

The process for obtaining a permanent labor certification has been criticized as being complicated, time consuming and requiring the expenditure of considerable resources by employers, SWAs [state workforce agencies] and the Federal Government. It can take up to two years or more to complete the process for applications that are filed under the basic process and do not utilize the more streamlined reduction in recruitment process.2

Struggling to deal with a backlog of more than 300,000 cases, the DOL has placed its hopes in PERM for the survival of the labor certification system. It will not be enough. PERM does nothing to resolve the internal contradictions that have plagued the labor certification system since 1965; until these internal inconsistencies are resolved, no procedural reform can achieve the desired effect.3 Whether PERM is a good or bad idea is not the issue. The issue is rather the extent to which PERM contradicts what Congress had in mind when it changed labor certification in 1965 to the scheme in place today. When Congress amended the INA’s labor certification language in 1965 from its original 1952 wording, what did not change was the responsibility of the Secretary of Labor, not the employer, to produce able, willing, qualified, and available U.S. workers at the place of employment for the advertised job opportunity. While the obligation of the employer to go forward and actually file an application did not exist before 1965, Congress never intended for the ultimate burden of proof to shift; at all times, both before and after 1965, that burden remained where it had always been, where the statute still places it, and where Congress has been content to leave it. If there is one misconception that has plagued the labor certification system since 1965, it is the conflating of the duty to file with the burden to identify and produce able, willing, qualified, and available U.S. workers. By treating them as if they were the same, the DOL has placed employers in the untenable position of having to prove a negative, something that they cannot do and something that Congress never asked them to do.

The system set up by the DOL after 1965 was exactly what Sen. Edward M. Kennedy (D-Mass.) had promised Congress when he served as the floor leader for this legislation: a system based not on individual recruitment but on statistical calculation.4 That is also precisely why the DOL lost case after case in the federal courts5: the willing requirement cannot be satisfied by statistics.6 Badly wanting an immigration bill that would abolish the national origin quotas and admit more immigrants, Sen. Kennedy agreed to the price set by organized labor-namely, a more stringent form of labor market control. Congress went along with Sen. Kennedy but did so in the belief that the Secretary of Labor would have access to the names of individual U.S. job seekers already on file with the state employment services, who were the human faces behind all these numbers.7

That is why the DOL placed the Foreign Labor Certification Program squarely within the Unemployment Insurance (UI) Division, now known as the Workforce Security Division. This was done so that the statistics would be readily available to the labor certification administrators at the DOL from the UI folks.8 Ultimately, the thought went, statistics represent people, and the states could funnel the names and addresses of such people to the Secretary of Labor who, in turn, would provide them to an employer so that labor certification would not be necessary.

Unfortunately, such plans soon fell victim to other budget priorities, and the money never materialized. Yet that was not the only reason that the DOL felt compelled in 1978 to turn to a system of individualized recruitment. It was not, and is not, just an issue of money but rather of speed. There was no way, without large amounts of money, that the SWAs could get the individualized information to the DOL in a timely fashion. If the information was not timely, then the employer would get an unreasonable decision. Today, the technology exists to do this very cheaply. The DOL resorted to bare statistics without the underlying individualized data files because there was no other way for the DOL to render a reasonably timely decision.

The DOL saw no alternative to this odious individualized recruitment process because there did not seem to be any other way around the plain language of the statute and no one believed that going back to Congress would resolve the contradictions. At some point, however, Congress will have to deal with the contradictions and PERM is precisely the transitional measure to hasten this day of reckoning.

It will not be pretty for Congress when that happens. The tensions that Sen. Kennedy dealt with in 1965 still exist. The notion of "willing" was a sop to the business interests whose agreement was necessary to secure passage of the overall 1965 Act. With the DOL in his pocket, Sen. Kennedy hoped to soften the impact down the road and he did so, as can be gleaned easily from even a cursory review of the early labor certification regulations.9 Only after this approach lost in the courts did the DOL reluctantly change its regulations; this resistance persists today. There was no malice or evil intent in all this. Sen. Kennedy did what he had to do to abolish the national origin quotas. In the process, he had to compromise on the language with labor.10 Such compromises, necessary for any complex legislation, inevitably lead to difficulties in implementation because of the resulting built-in ambiguities. This is a common experience in immigration legislation and that is why the agencies charged with implementing the INA historically have had a very tough time discharging their duties.

Whenever Congress debates any comprehensive reform proposal that potentially restricts employment, such as the Agricultural Job Opportunity, Benefits, and Security (AgJOBS) Act of 2003, which is the latest of many such examples, business insists once again, as it did in 1965, on bringing in an "availability" test, which is nothing more than a modern form of "willing." It is only the failure of such a test in the eyes of both business and labor that excuses the turn to immigrant workers. In this sense, both business and labor are united in moral opposition to immigrant workers. What distinguishes them is that business would carve out an individualized exception when the law is applied, whereas labor prefers an outright ban on any employment-based immigration whatsoever. This fundamental point has not varied from the early 1960s until the present day.

Legislative History: Present At The Creation

From 1885 to 1952, the Contract Labor Act prohibited employers from prepaying transportation expenses for, or otherwise assisting, foreigners who wanted to come to work in the U.S. The McCarran-Walter Act of 1952, which remains the foundation for U.S. immigration laws today, replaced this blanket preclusion with a more selective ban. As Demetrios Papademetriou and Stephen Yale-Loehr have noted,11 however, the 1952 version of labor certification was phrased in the negative-immigrants subject to it were automatically admitted unless the Secretary of Labor made a positive finding of able, willing, qualified, and available U.S. workers. This structuring was clearly intended to give the DOL the affirmative power to intervene to protect U.S. workers during recessions or in response to specific situations where the welfare of U.S. workers was endangered. But the initiative was with the DOL, and unless the agency interposed, an immigrant worker was admitted under the applicable country quota.12

How many negative rulings the Secretary of Labor made from 1952 until 1965 is in dispute. Rep. Peter W. Rodino (D-N.J.), who later gained national fame as the Chair of the House Committee on the Judiciary during the Watergate scandal, says that there were only six negative certifications between 1957 and 1961; by 1964, Rep. Rodino calculates that the DOL issued 18.13 Aaron Bodin, then chief of the Labor Certification Division, of the DOL’s Employment and Training Administration, reported in 1979 that there were only two strike-related exclusions from 1957 to 1965 with only 56 others during this same time. Most of these concerned low-skill occupations now on Schedule B, which PERM will likely abolish. Because such negative determinations did not extend beyond one job market in a specific city, they had little impact on safeguarding jobs for U.S. workers doing this same work who lived elsewhere.14

The reason the Secretary of Labor rarely acted is that he often had no idea there was a problem. There were only two ways for him to find out. Either a trade union complained that the alien was taking a job away from a member or a U.S. Consul received 25 or more applications during the year from a single employer.15 Charles Gordon, the dean of immigration lawyers then active, observed that pre-1965 labor certification was a "passive requirement to be fulfilled only if the Secretary of Labor demanded it."16 From Mr. Gordon’s perspective, the chief flaw in the 1952 construct was that it was rarely invoked.17

So what exactly did Congress change in 1965 and why? In a deceptively minor revision, the wording "if the Secretary of Labor has determined and certified" was changed to "unless the Secretary of Labor has determined and certified."18 A passive requirement rarely invoked becomes an active requirement often considered.19

Why did Congress act? Organized labor played a major role in the passage of the 1965 amendments. While in previous years labor had often supported restrictive (and ethnocentric) legislation, in 1965 the AFL-CIO lobbyists were lined up on the side of the White House and reform. The unions had not been satisfied with the operations of the negative labor certification program previously described, so they successfully sought a positive program in which the applications of aliens lacking familial or refugee credentials could be screened for their impact on U.S. labor markets. While strongly backing abolition of the national origin quotas, the AFL-CIO demanded a more robust labor certification as its price for support. So close was the partnership between the White House and the AFL-CIO that, at the time, the amended labor certification provision was colloquially referred to as the "Biemiller Amendment," for Andrew Biemiller, Legislative Director of the AFL-CIO. Congressional debate on this issue makes the point clear. All of the leading congressional sponsors took pains to emphasize that big labor was on board:

Sen. Kennedy: "I have already described the more stringent controls that this bill gives to the Secretary of Labor to insure against any adverse effects of immigration on U.S. labor. I would also point out that this measure has the complete support of the AFL-CIO; support that would not be forthcoming if the fear of job loss for Americans were real."20

Rep. Michael D. Feighan (D-Ohio), Chairman of the House Subcommittee on Immigration: "The AFL-CIO recommended that meaningful labor controls be incorporated into any reform immigration legislation and the new controls have the strong support of organized labor."21

Rep. Ray Madden (D-Ind.): "These new labor controls were recommended by the AFL-CIO and have the strong support of labor."22

Rep. Walter Moeller (D-Ohio): "It is significant that one of the strongest supporters of the bill is the AFL- CIO."23

The 1965 amendments were also designed to reassure skittish skeptics in Congress who wanted some mechanism to help control Western Hemisphere immigration.24 The Commission for Manpower Policy gives us the big picture:

It was also felt by others involved in the process that the labor certification provision could be used to limit Western Hemisphere immigration and that it would not be necessary to have a numerical limit on such immigration. The Senate was adamant about the 120,000 limitations, so the resulting legislation incorporated both labor certification and the Western Hemisphere numerical ceiling.25

Ironically, the Western Hemisphere cap did not become operative until 1968, while labor certification applied only to non-preference aliens who were not parents, children, or spouses of U.S. citizens or lawful permanent residents.26

All schoolchildren learn at their mother’s knee that the purpose and effect of the 1965 revision to labor certification was to take the monkey off the back of the Secretary of Labor and place it squarely on the broad shoulders of the employer. Right? After all, the experts tell us, did not Sen. Kennedy say as much on the floor of the Senate? Indeed, is that not what all defenders of the labor certification regulations, starting with Pesikoff v. Secretary of Labor,27 point to in support of their position, which has long since assumed iconic status? Well, as Al Smith, the famous progressive governor of New York State during the Roaring Twenties, used to say: let’s look at the record.

One must presume that Congress wanted to do something new in 1965, and it is certainly true that Sen. Kennedy spoke to this on the floor of the Senate on September 17, 1965:

Under current law, aliens who enter to seek employment are excluded from the country only if the Secretary of Labor has determined that their presence would have an adverse effect on the unemployment or the wages and working conditions of American citizens.... The new bill reverses this procedure. It places the burden of proving no adverse effect on the applying alien.28

This is not the only thing that Sen. Kennedy said, however, nor is it the most complete expression of his views on the subject. He went on to say something that shows why the pervasive micromanagement under PERM cannot be reconciled with what Congress thought it was doing in 1965:

It was not our intention, or that of the AFL-CIO, that all intending immigrants must undergo an employment analysis of great detail that could be time consuming and disruptive to the normal flow of immigration. We know that the Department of Labor maintains statistics on occupations, skills, and labor in short supply in this country. Naturally, then, any applicant for admission who falls within the categories should not have to wait for a detailed study by the Labor Department before his certificate is issued... [W]e would expect the Secretary of Labor to devise workable rules by which he could carry out his responsibilities under the law without unduly interrupting or delaying immigration to this country. The function of the Secretary is to increase the quality of immigration, not to diminish it below levels authorized by the law.29

In fact, contrary to what most of us have been taught to believe, it was the opponents, not the advocates, of the 1965 amendment who most accurately predicted what was to come. Sen. John Tower (R-Tex.), the Republican college professor who replaced Lyndon Johnson in the Senate, directly challenged Sen. Kennedy’s claim that the new labor certification procedure would not create an undue burden for the DOL:

Frankly, I see nothing but a tremendous bureaucratic nightmare in attempting to put these provisions into effect...the enactment of these provisions would, in my opinion, [place] an impossible burden upon the Labor Department if it is to administer them effectively. Imagine, if we will, the involved decisions that would have to be made in applying these restrictions to thousands upon thousands of prospective immigrants each year. In my opinion, the provisions are utterly unworkable and give every indication of being inserted in the bill merely to provide an answer to those who would raise questions concerning the bill’s effects on our labor market.30

Sen. Kennedy was not alone in downplaying the degree of difficulty that the DOL or the private sector would experience in adapting to this brave new world of labor certification. David North, whose study was funded by the DOL, remarked that the Senate report accompanying the legislation took "a cheerful, almost casual, view as to the ease with which a labor certification program could be initiated."31 Such deliberate insouciance was necessary to get the 1965 law enacted. The legislative process frequently demands that difficult points of implementation be quietly ignored when enacting an important principle. The 1965 law revision of labor certification was no exception. While it is true that the Senate Committee on the Judiciary spoke of the new system as a "reversal" of the old, it went on to assure all concerned that this was no big deal: "The Department of Labor should have no difficulty in adapting to this new procedure"32 because the "Department, through its Bureau of Employment Security and affiliated State Employment Security Agencies presently determines availability of domestic workers and the standards of working conditions."33 Precisely for this reason, the Senate felt that "there is no apparent need to increase facilities."34 That is why labor certification was placed within the UI Division: the program should be where the statistics already were.

Unfortunately, this congressional intent on the statistical basis for DOL determinations on labor certifications ran head-on into the unyielding language of the statute itself: "any excludable, unless the Secretary of Labor...has certified that...there are not sufficient workers, who are able, willing, qualified...and available."35 There is no indication here of a statistical process. This provision applies to "any alien"; i.e., any individual, not a statistical agglomeration of intending immigrants. Representatives in the House repeatedly made the point that the process would be an individual one:

Rep. H. Allen Smith (R-Cal.): "The Secretary of Labor, under the language in the bill, will be required to make a finding in the case individually that immigrants will not take a job for which there is a willing American worker nor upset the wage scales in the area."36

Rep. Madden: "New labor controls are established on the admission of all immigrant worker classes. These new controls require the Secretary of Labor to make an affirmative finding on an individual case basis. The job the immigrant worker will fill in the locality to which he is destined must be one where there is no willing, qualified, and available American worker."37

Rep. Feighan: "These new controls require the Secretary of Labor to make an affirmative finding on an individual case basis that, with respect to the job the immigrant worker is to fill in the locality to which he is destined, there is no able, willing, qualified and available American worker to fill that job."38

At the inception, therefore, there was an irreconcilable conflict between the statistical and individual nature of the labor certification process. This was to come back to haunt the DOL later. The most intractable contradiction, however, was that labor certification was set up to find something that did not exist. In other words, the DOL was being charged to find a negative fact. Unfortunately for the agency, and for Congress, there is no such animal. It is much like a man being told to prove that he is not guilty. Such a principle is grossly unfair and manifestly contrary to common sense. In order to trust a system, the people in it must believe that it is fair and has integrity. Fairness and integrity are positive virtues. No one can have confidence in a system based on a negative. The negative provision of the 1965 Act removed any hope of fairness and integrity in the labor certification system. That is why today all sides in the labor certification debate agree that no one believes in the system. The system was misaligned from the beginning and keeps getting worse. Much like the alignment of a car, if the misalignment is not corrected, it will keep getting worse. So, indeed, it has. PERM does not attempt to correct this imbalance.

At least one representative saw the problem early:

[T]he labor clearance clause is not a workable provision.... Can you picture the Department of Labor deciding whether or not an applicant plumber from Bolivia will displace any workers in Buffalo?...This provision directs the Labor Secretary to do the impossible.39

Simply stated, before 1965, the Secretary of Labor had to identify those occupational categories with a surplus of U.S. workers. There was no need for the employer to do or file anything. After 1965, the Secretary no longer had to be proactive. While the employer now had the burden of applying for the benefit, there was no indication that Congress ever intended for the employer to discharge this burden by a pattern of individualized advertisement and recruitment designed to prove that able, willing, qualified, and available job applicants did not exist. In perhaps the most perceptive article ever written on labor certification, in the author’s view, Harish Singhal explained why the burden of proof and the burden to apply are two very different things:

From the Secretary’s viewpoint, therefore, all that the new procedure achieved was to relieve him from the burden of having to act on his own initiative. The two burdens, the burden to apply for the labor certificate and the burden to determine whether the condition specified in section [212(a)(5)(A)] have been met, are separate and independent.40

What we tend to forget is that labor certification is a ground of inadmissibility. While the labor certification statute presumes that the alien should not be admitted into the U.S., Dr. Singhal noted that "there is no presumption that American workers are available or that employment of aliens would have an adverse effect on domestic wages."41 Dr. Singhal reminded those who had forgotten that the certification by the Secretary of Labor is a determination under the Administrative Procedure Act (APA).42 The APA provides that "except as otherwise provided by statute, the proponent of a rule or order has the burden of proof."43 Because INA § 212(a)(5)(A)44 nowhere provides that nonavailability will be presumed or established by the alien or the employer, Dr. Singhal concluded that the "burden [to determine whether the condition has been satisfied] clearly rests with the Secretary."45 Skeptics should remember that Sen. Kennedy did not want the new system of labor certification to diminish immigration but to increase it. That being the case, placing the burden of proving unavailability upon an individual employer who did not have access to all the data already collected by state employment services around the country but had to rely solely on his own vastly more finite resources would not only be inefficient public policy but, perhaps even more importantly, would contravene the very reasons why Congress changed the law in the first place.

The Department of Labor and its allies in organized labor have never accepted what PERM now seems all too ready to embrace: that a highly complex system of individualized recruitment is either the most faithful expression of legislative intent or its inevitable and most complete expression. In fact, they have consistently preferred what President George W. Bush has now put forward in his guest worker proposal-namely, a job registry-like approach based on reliable occupational statistics sent by the states to Washington, D.C. that allows the U.S. government to match a willing employer with eager workers while monitoring the courtship to make sure that both parties play by the rules. What the administration wants now is nothing more than a high-tech version of what the DOL has always preferred. Even while the 89th Congress was considering what to do about labor certification, Andrew Biemiller, the AFL-CIO Legislative Director, called for the "proposed establishment and up-to-date maintenance of a list of occupations and professions in which the skills, talents, and technical competence are most needed in the United States and for which a continuing need is anticipated."46 Mr. Biemiller felt that such a list of occupational and professional shortages, the same kind of information that President Bush thinks is important today, should be compiled by the Attorney General after consultation with the Secretaries of Labor, State, and Defense.

Rather than having to create any new government agencies, the Labor Department need only look to the expertise already available in the states that could be collected, analyzed and organized by the DOL’s Bureau of Employment Security.47 Only if we remember that labor certification was a statistical attempt to calculate occupational availability does it make sense to locate the program within the UI Division; the structure of the program tells us how the DOL thought it should operate and what its true purpose was.

When the Immigration and Nationality Act Amendments of 1976 extended the preference system, which had been in effect for the Eastern Hemisphere, to the Western Hemisphere as well, representatives of the DOL seized the occasion to testify in support of fundamental labor certification reform.48 Knowing full well, even then, that labor certification was not working for the benefit of unemployed U.S. workers, that most immigrants working in the U.S. came here through family ties without any labor market controls, that most immigrants who did use labor certification to come soon left their certified jobs for better opportunities, that the occupational distribution of immigrants was the same after 1965 as it had been before, Ben Burdetsky, then Deputy Assistant Secretary of Labor for Employment and Training, dismissed labor certification as "a costly, aggravating process which affect[s] a very small percentage of the immigrants who enter the labor force."49 The DOL had what it thought was a much better idea, namely a return to the pre-1965 negative version of labor certification:

Our proposal essentially would admit aliens as immigrants for employment within prescribed quotas and without individual labor certifications unless the Secretary of Labor acts to stop them. Such determinations would be initiated by the Department of Labor itself based on its monitoring of the supply-demand of occupational classifications by geographic area. They would also be initiated in response to information from any public or private source reasonably documenting that an adverse situation may result from the admission of aliens in an occupation and labor market area. A statutory alert mechanism would go into effect if a single employing establishment petitions for more than 25 aliens within 1 year. In the latter two instances, the Department would make a preliminary finding, using its national, regional and State employment service resources.50

When asked by Sen. Hiram Fong (D-Haw.) how this new-old system of labor certification would work, Deputy Assistant Secretary Burdetsky went back to almost precisely what Sen. Kennedy had in mind a little more than a decade earlier:

We would conduct a wide range of labor market studies. We have, you know, within the United States employment system that we work with, Senator, we have 2,500 local offices. Each State has a system. They do a lot of labor market analysis right now. We would arrive at decisions about certain occupations and certain places for which certification should not be allowed; in other words, we would not enter into it unless the Secretary determined that a certain occupation in a certain place should be certifiable. We would not look at each case on an individual basis. If that attempt by the Secretary does not appear, we would not enter into the scene at all.51

The AFL-CIO objected to this attempt to restore the status quo ante and Congress declined to act. Even while refusing to go back to the future, however, Kenneth Meiklejohn, Legislative Representative of the AFL-CIO, conceded, in response to a direct question from Sen. Fong, that the 1965 amendment did not require an individual certification. "We did not think it was necessary," Meiklejohn opined. An incredulous Sen. Fong could not believe what he was hearing, so he asked the witness to repeat his answer. Back it came: "We didn’t think it was necessary."52

So, if both the DOL and big labor realized that a general certification would suffice, why did we wind up with PERM? Why did the Congress and the DOL not follow through on the pragmatic approach that could have avoided the administrative burden of issuing individual labor certifications? As with most things in life, it came down to money, or rather the lack of it, as increasing unemployment, a mountain of new Great Society initiatives, the voracious appetite of the Vietnam conflict, and the Oil Shocks of the 1970s combined to gobble up the funding that otherwise might have allowed rationality to prevail. "Now, we certainly would like to have been able to do this," Aaron Bodin admitted in 1979, but "the state employment service agencies were already staggering under the new responsibilities heaped on them by the Great Society programs, and the resources simply were not there even though we know the workers were."53

Devil In The Details: What Was In The Regs?

The change from a negative to a positive form of labor certification acted as a sharp and immediate stimulant on the volume of labor certification traffic. Years later, Aaron Bodin remembered how anxious employers rushed in to file cases in unprecedented numbers:

The U.S. economy was booming in 1965 and applications for labor certification poured into the Department….In those early years, a high percentage of applications could be approved because qualified U.S. workers were generally employed and were unavailable for new job openings… Because of the expanding economy, the Department was able to reduce its workload somewhat through the adoption of a pre-certified list of occupations in short supply nationally, which we called Schedule A…The Department also issued a Schedule C list of occupations in short supply in some, but not all, areas of the United States and the Schedule B list…The Schedules are, I believe, conceptually sound and represent a pragmatic approach to the administrative burden of issuing individual labor certifications.54

There was no mention of individualized recruitment in the proposed labor certification regulations on November 19, 1965,55 or the final version of these same implementing rules that came out on December 3, 1965.56 There was no sense that employers had to advertise; the availability of U.S. workers, or their nonavailability, was based solely on statistics as embodied in Schedules A and B, respectively. Because these were the first regulations under the new regime, surely if Congress had intended to impose the same kind of evidentiary burdens on employers that exist today, the DOL would have spelled this out in the body of the regulations themselves; instead, they are conspicuous by their absence. When the DOL revised Schedules A and B the following year, once again the DOL could have articulated what employers had to do and when, much as in PERM. Once again, there is not even the faint hint of such an obligation.57 1965 had come and gone, but the supremacy of labor market data to determine availability went unchallenged.

In 1967, when the DOL first introduced the concept of regional, though not national, occupational shortage as embodied in Schedule C, there was no discussion of individualized advertisement or personalized recruitment.58 Not until February 1971 did the employer have an opportunity to file an individual labor certification if the advertised job was not on Schedule A or B.59 Even then, the regulations did not demand that the employer prove the absence of available U.S. workers; indeed, once again, as always, the DOL preferred to rely solely on labor market data as the factual predicate for its determinations. For the DOL, however, there was no contradiction between individualized determinations and statistical labor market employment data. The data were always backed up by the names and addresses of actual Americans in actual occupations who were currently unemployed; that is why they were collecting their unemployment checks from the state job services and that is how the states knew about them, and compiled the statistics on which the Secretary of Labor would then rely.

The concept of “willing” job applicants is also nowhere to be found in these regulations. When articulating the purpose of the 1971 rules, the DOL spoke only of certifying to the legacy INS that “qualified U.S. workers are not available.” Here the DOL was being faithful to what Sen. Kennedy wanted in 1965: throwing a sop to business by keeping “willing” in the statute but minimizing the practical impact of such a concession by leaving it out of the regulations. Nor was the omission of “willing” from the 1971 regulations an aberration; the omission was deliberate. In the 1975 regulations, when the DOL next had a chance to tell employers how the new world of labor certification would operate, it is apparent that the agency wanted only enough information to be able to assure the legacy INS that “qualified U.S. workers are not available and that [the alien’s] employment will not adversely affect wages and working conditions of the workers in the United States similarly employed.”60 Because it is manifestly impossible to identify a “willing” job applicant through statistical analysis, the DOL could not prove he or she existed and the employer did not have to prove the contrary.61

The “willing” requirement did enter on the scene in January 1977, after the DOL had taken its lumps in a series of court defeats that a later section of this article will discuss. The requirement came when Congress refused to return to the pre-1965 version of labor certification as the DOL wanted, or to take any other positive steps to resolve the inherent contradiction of how the Secretary of Labor could discharge his burden of proof while remaining faithful to congressional intent.62 It is only when the DOL could think of nothing else that the agency took the labor certification regulations away from volume 29 of the Code of Federal Regulations, where they had until then resided, and moved them to volume 20, which remains their home to the present day.63

Proving (and disproving) the negative soon caused both the DOL and employers difficulties. DOL officials could not come up with a reasonable proof whenever their negative finding was challenged by employers. To be reasonable in a negative context is a contradiction in terms. Naturally, then, the DOL came in for criticism from those affected by the effects of this contradiction:

The procedure to enforce this requirement has been an administrative nightmare. The decisional criteria are obscure and often unrevealed and the administrative officials frequently have been criticized for processes and determinations that are arbitrary and lacking in due process.64

Arguments flew back and forth, and so, as a practitioner remarked: “It has tripled the paper work.”65 When employers sought administrative review of labor certification denials, the DOL took to issuing form responses of unavailability based on statistics often not germane to the specific individual application before the DOL. The result was summed up by a practitioner as follows: “Administrative review of initial labor denials is often a sham, and judicial review an increasing necessity…. The labor certification program has returned the immigration practitioner to the dark ages of administrative law.”66

Had Congress agreed in 1976 to allow the DOL to avoid the need for individual labor certification and allow aliens to enter within prescribed numerical limits unless the Secretary acted to stop them, we would not be waiting breathlessly for PERM today. The DOL would have been quite content to extricate itself from this unwelcome briar patch and monitor occupational supply and demand to prevent any adverse impact on wages or working conditions. The system we have today is the one that the DOL went to only because it had no choice—because Congress failed to act when it had the chance. Time after time, the DOL could have resorted to supervised recruitment to enforce the post-1965 labor certification scheme and, time after time, it did not do so. Only when employment and the economy finally fell victim to the economic dislocation resulting from the Vietnam War and the Oil Shocks of the 1970s could the DOL no longer run away. Only then, seeing no exit, and lacking any hope in congressional intervention, did the agency take the first fateful steps down the road to PERM.

Though the DOL clearly addressed the issue of statistical versus individual that had bedeviled it in the courts, the 1977 regulations did not change the fundamental defect in the labor certification process: the fact that it was a negative process. PERM likewise suffers from this same fatal flaw. The labor certification system established by the regulations required the employer to “prove the existence of the non-existent.”67 Natural recruitment is not about proving unavailability, but the regulations require the employer to do the unnatural, the artificial. Most of the criticism of the labor certification process today is directed at the artificiality of the recruitment process. It should be recognized, however, that this artificiality arises from the process’s negative nature.

This unnatural system led to a number of unintended consequences. First, the interests of U.S. workers and alien workers were brought into conflict. This is the source of the concept of “diversion” wherein an employer can only hire either a U.S. worker or an alien worker (but not both) during the labor certification process. The resulting “win-lose” mentality now pervades the system. Parties forced into a “win-lose” situation look upon each other as adversaries, and thus trust, which is the glue of any system, is lost. It is no wonder, then, that we now have a system where the parties have lost trust in each other.

Second, the DOL rather than the natural parties (U.S. employers and workers) became the center around which the labor certification system revolved. Having employers and workers (both U.S. and alien) as a natural center would have led to a shared focus on increased economic output. By contrast, having the DOL as the artificial center distracted everyone from focusing on the economy. This is clearly seen today by the fact that all parties in labor certification care only about what happens to the labor certification—approval or denial—without giving much thought to what comes after.

Third, this artificial system, divorced from the dynamics of economic reality, was static. A static culture inhibits any structural reform, thus laying the foundation for a succession of crises. Moreover, each succeeding crisis will prove worse than the ones before. To return to the analogy of the alignment of a car, if a misalignment is not fixed, it keeps getting worse and the costs of repair keep going up. It is for this reason that labor certification seemed to lurch out of control from one crisis to the next. Disconnected from reality, the system kept sinking deeper into trouble with every attempt at getting out. Labor certification became a program without a purpose. As a commentator remarked at the time:

The labor certification program needs to be substantially re-evaluated. If the basic purpose of the program was to protect the United States labor market, the purpose has not been and cannot be achieved because only a fraction of the immigrants are required to obtain a labor certificate. If its purpose was to facilitate the needs of U.S. employers without creating a surplus labor force in competition with domestic labor, then the program has also failed, because, as implemented, it has disregarded employers’ needs and has made certification expensive and burdensome. It is time that Congress did something about the law or its implementation.68

Every attempt by the DOL to improve the efficacy of individualized recruitment has only served to illustrate further its inherent complexity and fundamental lack of coherence. PERM continues to treat labor controls only in the isolated context of a small proportion of total immigration. Yet it makes no difference to a U.S. worker whether he or she was affected by an alien who came in under a family- or employment-based category. It is logical, therefore, to consider labor controls within the entire context of immigration, both family and employment, rather than just a part. This is precisely what PERM fails to do. Once again, the DOL is going to take a bum rap. The truth is that no reform of the employment system can ever hope to work in the absence of a fundamental reordering of our national immigration priorities.

Soon after their introduction, the 1977 regulations began to have the effects that one would expect from a negative, unnatural, and artificial system. There was frequent disagreement between employers and the DOL on labor certification applications. Instead of finding their way into the district courts, however, such disagreements were now decided within the DOL. This was because employers did not have recourse to the federal judiciary without exhausting the administrative remedies the 1977 regulations had established.

Because the system lacked reality, it lacked any basis for reasonableness. Because the system was not reasonable, the DOL’s administrative decisions were neither consistent nor uniform. So the DOL revised the regulations to create a Board of Alien Labor Certification Appeals (BALCA) in 1987 to replace the system of appeals to single administrative law judges within the DOL. The rule creating the BALCA said, “[T]he Board will enhance uniformity and consistency of decisions.”69 A subsequent BALCA decision explained: “The purpose of the Board is to provide stare decisis for the immigration bar.”70 Subsequently, however, these goals were not achieved, and the BALCA invented a device (the en banc decision) to resolve inconsistencies in BALCA decisions.

The mere creation of the BALCA is symptomatic of the problem, but did not lead to a solution. Not only do BALCA decisions lack any precedential value, but they also betray a tendency to approach basic questions of interpretation from radically differing perspectives depending on the vagaries of the relevant facts in the cases at issue. Consider the question of alternative requirements, for example. In Delitizer Corp. of Newton,71 employers are told that such alternative requirements must be “sufficiently dissimilar” from the advertised job, but in Francis Kellogg,72 the BALCA tells employers that such requirements must be “substantially equivalent” to the advertised job. Which is it? Not only are the two cases irreconcilable, but their divergent approaches to a core issue leave employers confused and not knowing which way to turn. This makes intelligent business planning and coherent career development all but impossible. The inherent inconsistencies within the very concept of individualized recruitment not only defeats the BALCA’s well-intentioned attempts to unscramble the mess but actually turns them on their head so that, in the end, they only make the confusion that much worse.

If the BALCA represents one attempt by the DOL to smooth out the rough edges of individualized recruitment, Reduction in Recruitment (RIR) is another. Although a part of the labor certification regulations since the transition to 20 CFR, the RIR provision was generally unused until General Administrative Letter (GAL) 1–97.73 It is now popular to the point that many people today think that RIR is the only way to do labor certification. The exception has become the rule. The popularity of RIR is a testament to the failure of the micromanagement of recruitment because the recruitment report for RIRs is only the employer’s attestation. Under the terms of GAL 1–97 and its progeny, resumes for applicants need only be supplied at the employer’s discretion, and in fact need not even be retained in the records.

PERM moves 180 degrees away from these advances by micromanaging the recruitment process. The labor certification backlog reduction plan announced by the DOL on July 21, 2004,74 is the agency’s latest attempt to come to terms with the consequences of the 1978 regulations. The backlog reduction plan amends the regulations to permit the National Certifying Officer (NCO), who also serves as Chief of the Division of Foreign Labor Certification, to direct an SWA, or a regional DOL office, to transfer a pending labor certification application to a centralized processing facility. Essentially, the DOL wants it both ways: a system of individualized recruitment with the NCO having the power to shift backlogged cases to two processing centers, away from control by the states.

In essence, this “federalizing” of the labor certification system is designed not only to eliminate the backlog, but also to retain a system of individualized recruitment with the DOL calling the shots. No longer will dissatisfied state job services be able to threaten to suspend further processing of labor certification applications for lack of sufficient federal funding, as the Alaska Department of Labor pledged to do in July 1998.75 This was not the first, nor would it be the last, expression of state discontent with federal mismanagement of labor certification. The states inevitably tended to associate their interests with those of employers and industries in their area so that the program would work for the benefit of both business and labor. This could not help but conflict with a federal perspective that did not associate labor certification with the expansion of economic opportunity. Hence, the federal-state alliance was always grudging at best; at times, the frustration boiled over:

As State administrators we have considered many ways to expedite the handling of ALC cases and to reduce the tremendous backlogs. However, we need more support from USDOL to lobby for additional funds and seek regulatory changes. We have actually held informal discussions about returning both our ALC grants and the performance responsibilities to the Federal government...We dislike the idea of seeing our states return the ALC grant and are not ready to make that recommendation to our Governors. But we are unable to accomplish the job with inadequate resources and support. In addition, the negative impact this under funded program is having on our relationship with our state’s business community is causing serious damage. Your prompt attention to this request on the question of returning ALC grants and responsibilities to the Federal government is greatly appreciated.76

Some of the state job services even received bomb threats: “The frustration of the parties to the pending cases has resulted in bomb threats and other threats of violence to state agencies. Several states are considering whether to refuse to continue to operate the program under these untenable conditions.”77 While the backlog reduction rule does not give a timeline, DOL officials have stated publicly in a variety of forums that they expect the backlog to be erased within two years of PERM going live. By then, save for calculating prevailing wages, the states will have no continuing role to play. Long suspect for being too close to the industries in their respective states, and too willing to work with lawyers so that the cases get through, the state workforce authorities will now be relegated to the sidelines in PERM’s brave new world.

IMMACT 90: The Chance Not Taken

In 1990, Congress passed the Immigration Act of 1990 (IMMACT 90), Pub. L. No. 101–649, 104 Stat. 4978, in part to fill “the need of U.S. business for highly skilled, specially trained personnel to fill increasingly sophisticated jobs for which domestic personnel cannot be found.”78 Congress’s use of the word “increasingly” reflects its understanding of the dynamic nature of the economy and the consequent evolving nature of the jobs. By contrast, the Department of Labor (DOL) continued to look at jobs as static, a consequence both of its worldview and of the regulations the DOL had adopted for labor certification.

In IMMACT 90, Congress originally intended to carry out a wholesale revision of the labor certification process by replacing it with an employer attestation system, enforced through a complaint-based audit. As the House Committee Report explained:

The bill replaces the current labor certification process with a streamlined attestation process that places certain responsibilities on the employer and those U.S. workers who may be adversely affected by the importation of foreign workers. The current certification process neither serves business needs nor protects the domestic labor force. The Committee recognizes, however, the importance of the principles embodied in [INA § 212(a)(5)(A) (8 USCA § 1182(a)(5)(A))] which unfortunately have become camouflaged behind multiple layers of adjudication, imprecise recruitment guidelines, and lengthy processing time frames. The Department of Labor’s role is transformed from that of a processing instrument to a complaint-driven investigative and adjudicative agency, one which can serve as a labor resource for employers and workers alike. Moreover, providing the legal process for enforcement on challenges and complaints about attestation conditions gives meaningful protections for U.S. workers. Expedited time frames provide rapid turnaround to meet employer needs…With respect to the duties of the Department of Labor upon the filing of an attestation, the bill prohibits any pre-screening of the employer’s assertions and promises as long as the attestation is filled out in a complete manner. Instead, the resources of the Secretary will be directed toward investigating complaints received on the attestation…Attestations that have been challenged will go through a thorough administrative review procedure. This will focus Department of Labor resources on enforcement procedures, not on universal, routine and frequently unnecessary administrative screening and review activities.79

Congress here showed a keen understanding that the system suffered from a lack of trust, and that an unwarranted situation of adverse relations had been created between business and labor: “[T]he committee acknowledged that flaws in the current labor certification process had created a ‘lack of faith in the system’ and resulted in a ‘strain on the relationship between business and labor.’ ”80 By contemplating that employers whose attestations were not challenged would stand, Congress also demonstrated a practical way to bring the positive back into the labor certification system.81

Displaying further grasp of the positive, Congress would have made a positive complaint from an affected U.S. worker the basis of any DOL enforcement. Congress did not fully understand, however, that the problem arose from the process’s negative nature, still saying that “[i]n order to bring in an alien worker, an employer must file an attestation that it has made specific efforts to find and hire U.S. workers for the job without success.”82 Had IMMACT 90 included the originally planned wholesale revision of the labor certification system,we would not be in the predicament that we are in today. The obvious question is: why did Congress not act in 1990?83

The answer is that the DOL objected to it. A subsequent House Committee Report said:

[T]he Department of Labor is currently developing its own reform of the labor certification process which will incorporate labor market data to identify occupations in which foreign workers might be admitted into the U.S. without interfering with the job opportunities of U.S. workers. Secretary Dole’s letter to the Committee states her strong objection to this provision.84

The reform that Secretary Elizabeth Dole referred to here was the Labor Market Information Pilot Program, a provision that was included in IMMACT 90 and would have expanded a pre-selected list of shortage occupations kept by the DOL. This is known as Schedule A.85 Due to extensive negative comments received by the DOL, this promised reform did not ultimately happen. Notably, a Republican administration here objected to a Democratic Congress’s desire to reform labor certification. This only underscores the fact that the problems of labor certification are cultural and institutional, not partisan.

A golden opportunity to reform labor certification had been missed. Congress, once again, demonstrated its predilection of balancing the greater immigration quotas that IMMACT 90 made possible with greater labor controls. This time the greater labor controls took the shape of a numerical limit on H–1B visas, and the institution of a complaint-based audit system for H–1Bs. Unfortunately, the increased immigration as a result of both the Immigration Reform and Control Act of 1986 (Pub. L. No. 99–603, 100 Stat. 3359) and IMMACT 90 created a perception that Congress had left the system unbalanced, and without sufficient labor controls. Positing such a perception explains the subsequent crackdown by the DOL (and even the legacy INS) on employment-based immigration in the face of the prosperity of an economic boom with the low inflation and the low unemployment that the 1990s saw.

In 1993, Vice President Al Gore inaugurated the National Performance Review. From the point of view of labor certification, the impact of the National Performance Review was that the DOL, among other federal agencies, was asked to identify opportunities for reducing expenditures. The resulting savings would contribute to balancing the federal budget, which at that time was running heavily in the red. At the DOL, the labor certification program was a prime candidate for cutbacks. Being at the center of the labor certification process, the DOL experienced its artificiality and unreasonableness “up close and personal.” It would be only natural for budget officials and other high officials in the DOL to dislike such a program. With this background, the DOL began its “re-engineering” initiative for labor certification in 1995.86

The National Performance Review reported to the President in September 1995 that $223.8 million could be saved over five years in the labor certification program by “streamlin[ing] the alien labor certification process[,] by decentralizing authority to state employment agencies and automating form processing.”87 It described its proposal as follows:

Streamline Alien Labor Certification: Streamline and speed up the [DOL’s] Alien Labor Certification process by decentralizing authority to state employment security agencies, consolidating DOL regional processing centers from 10 to four, and automating forms processing. Under this proposal, DOL will conduct spot audits of about 2 percent of its cases rather than review all state certifications. Also, states will be authorized to charge user fees to those few employers who use this service.88

None of the proposed changes that the DOL outlined in the National Performance Review actually happened. What did happen was that funding for labor certification was reduced from $60 million in 1993 to $40 million in 1997. These cuts contributed to the National Performance Review’s aims, but none of the proposed reforms materialized. The subsequent rapid increase in the backlog of labor certification applications should not have come as a surprise. An immigrant worker trapped in that backlog best summed up the situation on an Internet newsgroup: “In order to balance government budget, we are those people to be sacrificed.”89

With the National Performance Review looking for federal programs to cut, the DOL’s Employment and Training Administration (ETA) requested the DOL’s Office of Inspector General (OIG) to conduct an audit of the labor certification program: “Since ETA requested the audit, we have launched our own reengineering efforts through the National Performance Review initiative to address program weaknesses and to achieve a more rational allocation of resources.”90 Herein lay a double irony: not only was the ETA asking for an audit that criticized the ETA, but the ETA felt alienated from its own roots because the ETA’s predecessor agency grew out of the Bureau of Immigration. Such cultural contradictions are the inevitable, though unintended, consequence of a system founded on a negative premise.

The OIG released its final report of the audit to the DOL in May 1996. This report has come to be called the OIG Report.91 The OIG Report severely criticized the labor certification program and alleged numerous “abuses” in the program. One commentator pointed out: “Many of the ‘abuses’ cited by the report are required by the system.”92 A later Senate Committee Report criticized the OIG Report and quoted a review:

The audit report is filled with a series of fundamental errors, including factually inaccurate statements of congressional intent, lack of knowledge of the Immigration and Nationality Act, seriously flawed statistical methodology and incorrect measurements […] sufficient to render the report completely invalid as a tool for policy makers.93

Much of the OIG Report’s criticism was directed at the recruitment program required under the labor certification process, which the Report labeled as a “paper shuffle.” Even critics of the OIG Report agreed that “[t]he biggest structural problem, from which many of the other problems flow, is […] a micro-managed, artificial recruitment campaign.”94 What both the OIG and its critics did not recognize is that a pointless paper-shuffle and micromanagement neccessarily result when a negative premise disconnects a system from reality.

Just as the DOL was displeased by the program, so were its users: “While the process may be a sham, it is hardly being manipulated and abused by employers. The sham is all the DOL’s making, not the result of employer abuse and manipulation.”95 The DOL was well aware of the artificial nature of the recruitment campaign, which lies at the heart of the labor certification program under the agency’s own regulations. Its own re-engineering questionnaire posed this question for comment: “How are the recruitment requirements of the permanent labor certification process contrary to the usual recruitment and hiring practices of employers?”96 The DOL’s attitude toward labor certification can be seen clearly in comments by ETA’s Assistant Secretary in Appendix II of the OIG Report:

[A]pproved cases represent about 49% of the total number of applications filed.…This overall approval rate indicates that the program preserves jobs for U.S. workers to some modest extent. Clearly, strengthening the program would greatly improve protection of U.S. jobs.97

In other words, the DOL was saying that a 51 percent denial rate indicates that the program works to a modest extent. The logical conclusion is that only a 100 percent denial rate would satisfy the DOL. This illustrates the “win-lose” mentality that characterizes the DOL’s approach toward labor certification. PERM looks at immigration as a zero-sum game in which every job that goes to an alien worker means one job fewer for a U.S. worker. Such hostility is the natural result of an artificial system based on a negative premise. Nor is the DOL alone in having negative feelings. In a blistering critique of the OIG Report, a commentator remarked: “The real issue is that employers…must wend their way through a dysfunctional system that they had no part in constructing and which is just as frustrating to them as it is to DOL.”98 The fact is that all parties involved in labor certification are frustrated, and all are dissatisfied with the program’s state. The root cause of the problem, however, has not yet been recognized. As the history of this program shows, the negative premise of lack of availability lies at the root of the system’s problems. Clearly, another way must be found to affect the labor controls that Congress intends.

Congress expressed a specific intent as to the time frame within which the labor certification system must operate: “[T]he Secretary has determined and the time of application for a visa and admission to the United States.”99 Common sense would indicate that what Congress meant by the words “at the time” was some reasonable period, such as 30, 60, or even 90 days. Surely Congress did not intend the multi-year time frames that are common experience today. Even the OIG Report criticized this aspect of the labor certification process:

Yet, on average the alien will not become a permanent resident for approximately one and a half years after the employer files the alien labor certification application.… [I]t takes on average 525 days from the time the employer first files an ETA–750, Application for Alien Employment Certification, with the SESA for DOL to approve the application and for INS to approve the petition (INS I–140) and adjust the alien to permanent resident status (AOS).100

The original version of IMMACT 90 showed that Congress was well aware that its expressed intent on time was not being fulfilled by the current labor certification process, which was one reason why Congress contemplated replacing the system with an attestation process:

Additionally, the current labor certification process required to prove that this search has occurred is fraught with time-consuming hurdles. In many parts of the country this part of the immigration process alone can take 15 months. The bill responds to this problem by allowing employers to file an attestation as to recruitment, payment of prevailing wages, and strike conditions, thus replacing the lengthy, adjudicatory, labor certification process with a process that allows attestations that have not been challenged to become effective.101

As Congress implicitly realized, lengthy time-frames are yet another unintended consequence of a negative system. Congress comprehended that a positive system—with attestation, and a complaint-based audit—would not suffer from this defect.

Another unintended consequence of a negative system is its high expense. As Sen. Kennedy remarked in 1996:

At present, the Department of Labor certifies an employer’s application for an immigrant worker based on a complex, labor-intensive, and expensive preadmission screening system.… The Commission on Immigration Reform estimated that labor certification cost[s] employers $10,000 per immigrant for administrative, paperwork, and legal costs.102

This excessive price tag has serious economic consequences, because it effectively deters U.S. employers from making use of the immigrant worker program that Congress has established. As a commentator notes:

[T]he [labor certification] procedure is so costly and time-consuming to employers that it undoubtedly discourages employers from hiring foreign nationals and seeking labor certifications for them. In fact, a number of major corporations have stated publicly that they will not seek permanent residence for any worker (some with exceptions for very specific cases). Some of the country’s largest companies, with tens of thousands of employees, file five or fewer labor certification applications per year.103

Contrast this statement with the DOL’s own acknowledgement that “the permanent labor certification program was established to help employers.”104 Clearly, Congress did not intend the program to reach this state.

‘Here Come The Judge’: What The Courts Have To Say

There is no more important case in the history of labor certification than Pesikoff v. Secretary of Labor (Pesikoff). 105 The facts were not in dispute. Dr. Pesikoff was a child psychiatrist in Houston, Texas. When the lawsuit began, his wife was a law student and they had two children, ages two and four. They wanted a live-in maid and the DOL said no, based on the availability of 189 U.S. workers registered as maids in the Houston office of the Texas Employment Commission. Adding greater weight to the case was the decision by the Pesikoffs to retain Jack Wasserman as their lead counsel.

All court cases from 1974 to the present day can trace their intellectual lineage to this landmark ruling. In a momentous two-to-one reading of the 1965 labor certification amendment, the Pesikoff panel placed the burden of proving the absence of able, qualified, willing, and available U.S. workers squarely on the employer, not the Secretary of Labor. "We think it clear," Judge Skelley Wright wrote for the court, "that Congress did not intend [section 212(a)(5)(A)] to create an employment placement office in the Department of Labor."106 In other words, the employer must document the absence of adverse effect; the Secretary of Labor is under no obligation to go out and find a qualified U.S. worker who is available and wants the job. Any other result, so the majority found, would be contrary to the intent of Congress in reversing the burden of proof.

Is that all there is to Pesikoff? Perhaps not. Several other nuggets can be found in the opinion upon careful examination. Interestingly, neither the DOL nor Mr. Wasserman contended that the regulations required Dr. Pesikoff to place the newspaper advertisements, consult the employment agencies, or contact the friends he used in an effort to provide a record on which the DOL could certify the job. In fact, nowhere do we ever learn what was wrong with this pattern of diversified recruitment. In denying the labor certification, the DOL based its decision entirely on the guidance from the local office of the Texas Employment Commission that there were 189 qualified and available U.S. workers. The fact that none were willing to live in did not matter because Judge Wright dismissed such a requirement as a mere preference not binding on the DOL or the court. This issue, not the question of what Dr. Pesikoff had to do to support his petition, was the seminal issue in the case. As Judge Mackinnon noted in his celebrated dissent, the "pivotal issue is the reasonableness of the personal preference characterization."107 Moreover, while Mr. Wasserman did raise the issue of non-availability, he did not contend that the DOL’s regulations went beyond the intent of Congress, nor did he argue that the 1965 Act did not shift the burden of proof away from the Secretary of Labor where the plain language of the law had always placed it. Judge Mackinnon implicitly made precisely this point in prophetic words that would echo down the years to the present day:

The majority would have the employer carry the ultimate burden of persuasion that he has been unable, or that it is impossible, to find an "able, willing and qualified" U.S. worker-in a sense, the majority would have him prove the existence of the nonexistent, a sometimes difficult proposition.108

Pesikoff never mentions the proper way to identify or locate a "willing" job applicant, a rather surprising omission for a case whose iconic status derives, in part, from its close reading of the relevant legislative history. Judge Wright seems to approach the notion of “willing” in the same common-sense fashion that Sen. Kennedy probably used when shepherding the 1965 amendments; namely, if somebody bothers to register with a state employment service in the occupation for which certification is sought, that, by itself, is willingness enough. Neither Judge Mackinnon nor Mr. Wasserman contended otherwise.

What is often overlooked about Pesikoff is the fact that there are many courts in many circuits that disagree with it. Pesikoff is justly celebrated but it has never been without its critics, then or now, both on and off the bench. Indeed, Judge Wright openly acknowledged as much by this forthright concession: "We are aware that some recent District Court cases make a contrary interpretation of the statute. See also First Girl, Inc. v. Reg’l Manpower Adm’r of U.S. Dep’t of Labor...; Digilab, Inc. v. Sec’y of Labor, D. Mass., 357 F. Supp. 941 (1973); Bitang v. Reg’l Manpower Adm’r of U.S. Dep’t of Labor, N.D. Ill., 351 F. Supp. 1342 (1972); Golabek v. Reg’l Manpower Adm’r, U.S. Dep’t of Labor, E.D. Pa., 329 F. Supp. 892 (1971)."109

Pesikoff turned not on burden of proof, but rather on the willingness of two circuit judges to categorize the live-in requirement as a mere preference that could be cast aside. Other courts have shown greater deference to what the employer thinks is important. The Fifth Circuit criticized this as an "Orwellian designation"110 and pointedly remarked that to use "denigrating language to transform an employer’s reasonable requirements into something less seems to us unfair."111 Similarly, the Eighth Circuit held that the "job requirements of an employer are not to be set aside if they are shown to be reasonable and tend to contribute to or enhance the efficiency and quality of the business."112 It would not do violence to Pesikoff to suggest that its relevance to contemporary labor certification is diminished somewhat by the factual setting in which the case arose.

Pesikoff’s insistence that the Secretary of Labor should not be expected to be a talent scout for the nation can be a double-edged sword for the DOL. If the Secretary is not obliged to produce able, qualified, willing, and available workers for an employer who needs, but cannot find one (the First Circuit reminds us), then the Secretary "should not have the privilege of determining the qualifications of any particular applicant for the job to be filled." Nor, without specific and articulable reasons, can the Secretary of Labor "have the right to attack the good faith of an employer’s personnel procedures."113 What Pesikoff gives the Secretary on burden of proof, it takes away in terms of the degree to which a Certifying Officer can substitute his or her judgment for that of the employer when deciding what core requirements a labor certification can and must contain. Also, as mentioned above, Pesikoff is a statistics-driven case. The court never orders the employer to advertise or recruit; Dr. Pesikoff chose to do that on his own.

If statistics won the day for the DOL in Pesikoff, however, the results were often less to the DOL’s liking in other cases and other judicial circuits beyond the District of Columbia. The Seventh Circuit, for example, consistently and firmly rejected what it derisively termed "the mere incantation of raw sufficient evidence of the availability of acceptable American workers."114 The courts have not been reluctant to question the reliability and probative value of the statistics upon which the Secretary of Labor relied both in Pesikoff and its progeny. Throughout the 1970s, when the Secretary’s numbers were repeatedly impeached and found wanting, the inevitable result was a reversal of the denial of labor certification. While Pesikoff was willing to order the employer to prove the existence of a negative, other courts have been more skeptical as to the value of such an exercise: "Since it is clearly not feasible for an employer to affirmatively and conclusively show that acceptable alternative American workers do not exist," the Sherwin-Williams court opined, "the denial of certification must rest on some meaningful statistics."115 If the Secretary of Labor declined to assume the burden of proof, he or she had better hope that the courts drew the same inference from the raw numbers as did the DOL. That is a dangerous game to play, one that the Secretary did not always win. As Bitang v. Regional Manpower Administrator116 bluntly reminded the DOL: "[T]he statute conferring upon the Secretary of Labor the responsibility for making these determinations certainly requires more than blind and unquestioning acquiescence in a state agency’s ultimate conclusions."

The statute has never varied in requiring the Secretary of Labor to make a finding as to the existence or absence of able, qualified, willing, and available U.S. workers. If the Secretary elects to reduce his or her administrative burden by relying on the necessary market data, then sure and swift access to such data is a necessary precondition to any timely decision. The whole regime of individualized recruitment was adopted by the DOL as a last resort in 1978, more than a decade after the 1965 amendments, precisely to get this information to the Secretary when no other way seemed to exist. If the Secretary could not get information from the states quickly enough that would be the factual basis for the discharge of his or her statutory responsibility, perhaps the employer could be compelled to do so as the price of certification. That is why Title 20 of the Code of Federal Regulations became the repository of the labor certification program.117

In truth, although not admitting so publicly, the Secretary of Labor could not deal with the unintended consequences of Pesikoff, which ironically turned out to be very much of a mixed blessing. Precisely because it made such broad, sweeping claims for the Secretary’s authority, Pesikoff elicited such a sharp and sustained judicial response as to compel the DOL to adopt sub silentio Judge Mackinnon’s dissent.118

20 CFR § 656 represents a tactical retreat by the DOL that leaves the Secretary with the ultimate burden of proof but mandates that the employer become an active junior partner in its discharge. These regulations exist not to protect U.S. workers or help the employer, much less the alien beneficiary, but to create a solid administrative record that would reduce the potential for arbitrariness and withstand judicial scrutiny by requiring the employer to amass the most reliable information possible on potential adverse effect for review and action by the DOL. As the Seventh Circuit noted in Industrial Holographics Inc v. Donovan,119 requiring the employer to advertise and recruit as part of its burden of production is a preliminary step that does not transfer or "improperly shift" the burden of proof away from the Secretary of Labor. The employer’s recruitment efforts do not relieve the Secretary of Labor of his or her her burden but rather "enable the Secretary to make an informed decision based on reliable evidence."120 It is the absence of such reliable evidence that persuaded a federal court to reverse the Secretary’s decision and direct the issuance of a labor certification pursuant to the Administrative Procedure Act.121

Conclusion: Where Do We Go From Here?

Fundamental change only occurs at those rare moments when all the players believe it is in their interest for it to happen. Now is such a time. It may not happen again for a long time. History and our own conscience will doubtless judge us harshly if we do not act. The volume of labor certification traffic has placed such a severe strain on the system that its internal contradictions can no longer be tolerated.

As noted earlier, the basic reason why U.S. immigration laws do not protect U.S. workers is the total absence of any labor controls on family-based migration that accounts for the majority of those who come here each year. Labor certification cannot be changed unless the employment-based system is also overhauled. Such a reform necessarily would require a radical restructuring of the family preferences to emphasize the integrity of the nuclear family and eliminate chain migration. The U.S. should no longer tolerate visa backlogs that divide the families of lawful permanent residents for years at a time. The spouses and unmarried minor children of all such green card holders should be treated as immediate relatives and be able to come into the U.S. without any numerical restrictions. All other family-based categories should either be abolished, with their numbers transferred to the employment categories, or such migration should be subject to labor certification. This would not prevent the extended family members from immigrating to the U.S., but it would require a proper system of labor controls to be in place as a way to promote the national interest and protect the legitimate interests of the domestic work force.

No one can defend the current system of labor certification with a straight face: "Most observers agree that the present system, which costs taxpayers upwards of $65 million a year, does little to safeguard jobs, wages or working conditions for U.S. workers."122 It does not protect U.S. workers because most people can come to work permanently in the U.S. without having to be certified. It does not help employers to run their businesses more efficiently so that they can expand jobs in the U.S. or confront competitors abroad. It does not help the aliens to develop their talents to their normal and maximum level, but rather serves to frustrate such development by essentially placing their careers on hold for several years while the torture of labor certification plays itself out. It does not help the general public to understand how immigration is in their best long-term interest. It does not facilitate the creation of a strong and sustaining national consensus that places employment-based immigration at the heart of our national economic strategy to confront the global challenges of a 21st-Century information age economy. Everyone who touches labor certification today is diminished by the association, none more so than the DOL itself. It frustrates the DOL and forces it to divert invaluable time, talent, and resources from its core mission of protecting U.S. workers:

The required labor market test is not designed to survey the labor pool available at the time the alien actually adjusts to permanent resident status. The PLC program requires employers to test the labor market for the availability of able, willing, and qualified U.S. workers at the time of application for a visa and admission to the United States on the alien’s behalf. Yet, on average the alien will not become a permanent resident for approximately 1½ years after the employer files the alien labor certification application.... [I]t takes on average 525 days from the time the employer first files an ETA-750, Application for Alien Employment Certification, with the SESA for DOL to approve the application and for INS to approve the petition (INS I-140 [Immigrant Petition for Alien Worker]) and adjust the alien to permanent resident status (AOS). The labor market test conducted at the beginning of the application process does not identify the pool of qualified, willing U.S. workers who may be available at the time the alien adjusts to permanent resident status.123

The DOL, after all, did not want to wind up in a maze of contradictions with no exit; a series of stinging judicial defeats left it with no choice:

In the mid 1970’s, the Courts interpreted the term "willing" in a manner that precluded ETA from using labor market information to make such determinations. Consequently, DOL has been required to show that there is a specific U.S. worker who is qualified and able to accept the employer’s job before an application can be denied. ETA’s operating experience and your report show that U.S. workers have little or no chance of being hired for jobs in which aliens are already employed. Therefore, it hardly seems reasonable to put U.S. workers through the frustration of applying for jobs which are simply not available to them.124

Having to operate without effective congressional oversight, and unable to clarify the ambiguity of the 1965 amendments by frequent resort to the regulatory process, however stringent the interpretation, the DOL did not know which way to turn: "The biggest structural problem, from which many of the other problems flow, is that a micro-managed, artificial recruitment campaign must be conducted by the employer at an arbitrary point in the real-world recruitment and employment cycle through which a typical employee moves."125 Whatever their interests were, employers, lawyers, aliens, and regulators were not well served by a regulatory regime weighed down by a command and control mentality that simply could not accommodate the need for rapid change in an economy that demanded it:

That the DOL has arbitrarily chosen a point in the hiring and employment process that bears no relationship to the actual time that the employer has recruited and hired the alien, thereby creating an artificial recruitment procedure, is not the fault of employers. In fact, employers are frustrated that they must expend the time and money to repeat a process that they have already completed. While the process may appear to be a sham, it is hardly being manipulated and abused by employers. The sham is all the DOL’s making, not the result of employer abuse and manipulation.126

The U.S. has never looked itself in the mirror and honestly decided what kind of an employment-based immigration system it wants and how much it is willing to pay to get it. The entire labor certification mess is a poignant illustration of what happens when immigration is seen as a political problem rather than an economic opportunity. The country needs a labor market control system that focuses not on proving the unavailability of U.S. workers, because it is impossible to prove a negative, but examines what concerns similarly situated U.S. workers the most; namely, whether their common employer is paying the alien worker the same wages with the same benefits as the U.S. worker. As long as this is taking place, it should not be necessary to show the utter inability to recruit a U.S. worker. In this economy, it is the market, not the government, that makes the most basic economic decisions.

The focus of PERM is all wrong. It creates bright-line distinctions that can be used to punish past transgressions. For this reason, its mindset is insular, that of a "fortress America" that looks only to preserve current employment rather than, as it should, being interested in the creation of new jobs and the mining of rich, but untapped, veins of future economic opportunity. If we accept the premise that U.S. workers are best helped by a rising economy that strives to improve our collective living standard, then whether or not an employer can recruit a U.S. worker for a particular job is irrelevant. What counts is whether that employer can demonstrate how the certification of the advertised job opportunity filed on behalf of a pre-designated alien beneficiary makes economic sense. Does it create more jobs for U.S. workers both now and in the days to come? Does it enable the employer to hire more people by becoming more profitable? These are the questions that any system of labor market control must ask and answer.

If opponents of labor certification reform oppose doing away with verification of unavailability for all cases, let us suggest that only those employers with a certain level of labor certification submissions, perhaps modeled after the definitions of being an H-1B dependent employer found in the American Competitiveness and Workforce Improvement Act of 1998 (ACWIA), Pub. L. No. 105-277, 112 Stat. 2681, be required to demonstrate insufficient recruitment. Employers who are not "labor certification dependent" in this sense would not have to make such a showing. The labor certification should not be filed with the DOL, but as a companion document with the immigrant petition directly to U.S. Citizenship and Immigration Services (USCIS), which must conduct any audit within a predetermined amount of time. USCIS need not separately approve the labor certification. Approval of the immigrant visa petition, by definition, should be evidence of acceptance of the employer’s representations made at the time of submission.

Even if PERM made sense in theory, which it does not, it would not work because it does nothing to change the allocation of immigrant visas which, at present, bears little, or no, relation to what the U.S. needs. Why, for example, should all nations receive the same number of immigrant visas? If the concept of charging immigrant visas to the country of birth makes some sense in the context of family immigration, it loses any justification when it comes to the allocation of employment-based green cards. Because we issue these things to help ourselves create and preserve economic opportunity, what difference does an accident of birth make?

There is an easy and obvious solution, which is to award immigrant visas based not on place of birth but on occupational category. The very notion of limiting immigrant visas by country has no place in the global economy of the 21st Century. It is an idea whose time has come and gone. To strengthen the U.S. economy, the focus of any labor control mechanism should be to attract and retain essential workers for a wide variety of jobs in both the old and new economies. Talent, not national origin, should count.

This seems like a rather bad time to argue for a campaign to enact a more enlightened immigration policy that would eliminate labor certification, rationalize our system of labor market controls, and make immigration a cornerstone of our international competitive position in the global economy. Yet, our failure to do so is a prescription for economic stagnation and societal decay. The American people might or might not accept an enlightened immigration policy; they have never been asked. Once they have an immigration law they can believe in, the never-ending series of crises and emergency campaigns can finally be replaced by a sober examination of what the nation needs and what it is willing to do to satisfy such needs. There are so many new laws and regulations that observers of good will and keen intellect legitimately feel overwhelmed. It is precisely when the headlines scream the loudest that we must step back from the moment, take a deep collective breath, and quietly set about the task of doing something for our clients and our country. The fact that fundamental change is difficult, even painful, only suggests how much we need it.

  • Effects of 9/11 on the Nation’s View of Immigration and Labor Certification

Now is not the time to be defensive, but outspoken. Immigration was essential for the U.S. economy before the terrorist attacks of September 11, 2001 (9/11), and remains so today. Our immigration policies were a drag on job creation and economic growth before 9/11 and remain so today. The legacy INS and the DOL thought of immigration solely as a political problem rather than an economic asset before 9/11, and their views are still the same. Business and labor wanted to get rid of an employer sanctions system that frustrated employers and terrified aliens, and they still do. Immigration was and is the only effective answer to the "graying of America." It was and is the only effective strategy to revive decaying urban neighborhoods and restore our largest cities to life. These fundamental realities have not changed and, now more than ever, the nation needs to be reminded of them. We cannot draw closer to the world in order to fight terror while pursuing an immigration policy based on a "fortress America" model.

The 9/11 attacks carried out by noncitizens have many talking about the difference between immigration and terrorism, and the need to make changes to our immigration system. America’s immigration system can no longer operate as it did before. PERM is such a crazy-quilt collection of unrelated policies and procedures that no one can understand it or make it work coherently to achieve a common set of objectives. It does not take a prophet to figure out what will happen if PERM goes live. The DOL will be forced to pile regulations on top of regulations so high that even the most informed experts are beaten down by the weight of the law. Such complexity has its cost. Employers will then spend time and money, precious commodities that would be better spent to hire new employees rather than navigating a whitewater river of procedural rapids. Human resources professionals will complain, employees will soon become bewildered and even the lawyers who see in such a thicket promising opportunities for continued business will give vent to their angst. In fact, lawyers should be the most stalwart champions of simplicity because a more rational system would encourage more participation from a new infusion of clients who have traditionally been intimidated from venturing into what they regard as a Byzantine world where logic is suspended.

The pro-immigration advocates point out that the enemy is not immigration, but terrorism. True enough, but because the immigration regime is impossible for the public to understand, this becomes a distinction without a difference in the popular consciousness. Until 9/11, all parties to the debate played an "inside-the-Beltway" game for several reasons. First, they were the only players. Second, they all spoke a common language based on shared knowledge of existing complexities. Third, the more ins and outs there were, the more change could be regulated and controlled so that any reform, by definition, would be incremental rather than systemic. Fourth, knowledge was power, and lack of knowledge among the public as a whole meant that the institutional alignments in Washington, D.C., were never seriously threatened. Fifth, and most importantly, immigration was neither conceived of nor administered as a strategy to bolster homeland security, so that the need for informed civic understanding never came to the forefront of the emerging national debate.

The 9/11 attacks changed all this utterly. In a democratic society, the logic of any successful national policy must be transparently obvious to those who have to obey and support it. That is why complexity for its own sake is not only of little benefit to its intended beneficiaries, but actually frustrates any coherent attempt to make the system more amenable to consistent interpretation and effective enforcement.

  • Why PERM Will Fail

PERM will fail, if adopted, because it ignores the imperative of simplicity. We must not be afraid of simplicity. It can be just as easy to understand or put into practice as its more complex cousin. A law that can be understood will be invested by the good sense of the American people with the moral legitimacy that our present system so manifestly lacks. For this reason, even if there is sharp and fundamental disagreement on immigration policy, that should not prevent us from putting a new immigration scheme into place. Indeed, the very fact that so many people feel so strongly about so many different aspects of immigration law actually highlights the need for a grand conclave that can give all concerned parties a chance to be heard. The American people must understand the law and feel they have a stake in its interpretation, enforcement, and evolution. It must belong to them-not to lobbyists, lawyers, bureaucrats, or scholars. Immigration is not international social work but an enlightened exercise in national self-interest. Immigration exists to serve the nation, not the other way around. We must use immigration the way we use tax policy, interest rates, or trade restrictions: to make us a more prosperous people.

Opponents of immigration should no longer be able to frustrate immigration through arbitrary or capricious enforcement that clearly goes beyond what Congress intended. The U.S. needs much tougher enforcement and higher levels of immigration. At the same time, supporters of immigration must stop acting as if 9/11 never happened. They have nothing to fear from honest enforcement motivated by a sincere desire to protect, not to punish. The real threat to U.S. workers comes not from the distinctly limited number of employment-based immigrants but from the much larger number of family-based immigrants whose entry is unchecked by any labor market control. Employment-based immigration should be focused on the creation of new economic opportunity rather than the preservation of what exists now. Growth, not protection, should be the goal. Facilitating future growth, not punishing past transgressions, is what any labor certification system should be all about.

PERM continues to misallocate the burden of proof and place it where Congress never intended for it to reside: on the employer. Consider the burden of proof language in the proposed PERM rule:

If the Secretary, through ETA, determines that there are no able, willing, qualified, and available U.S. workers and that employment of the alien will not adversely affect the wages and working conditions of similarly employed U.S. workers, DOL so certifies to the INS and to the Department of State, by issuing a permanent alien labor certification. If DOL cannot make one or both of the above findings, the application for permanent alien employment certification is denied. DOL may be unable to make the two required findings for one or more reasons, including: (a) The employer has not adequately recruited U.S. workers for the job offered to the alien, or has not followed the proper procedural steps in 20 CFR part 656. (b) The employer has not met its burden of proof under section 291 of the Immigration and Nationality Act (INA or Act), [8 USCA § 1361], that is, the employer has not submitted sufficient evidence of its attempts to obtain available U.S. workers, and/or the employer has not submitted sufficient evidence that the wages and working conditions which the employer is offering will not adversely affect the wages and working conditions of similarly employed U.S. workers.127

The DOL appears to realize that it has no statutory authority in INA § 212(a)(5)(A)128 for asking the employer to prove anything. Hence, in its proposed PERM rule, the DOL clutches at the straw of INA § 291129 as authority to ask the employer for proof. Section 291 binds aliens to produce proof, however, not employers. The plain words of § 291 illustrate that it does not apply to labor certification. A statute that speaks of proving the eligibility of an alien to enter the U.S. is a thin reed upon which to rest the PERM rule. At first, this appears to be a minor point, and it is. Yet, step back a moment and take in the full picture-the mindset that animates PERM is a continued belief that the employer must prove a negative before the labor certification can be issued. Whatever one thinks of this notion, one thing is certain: that is simply not what Congress thought it was doing in 1965.130

The real problem with PERM is not only that it goes beyond what Congress thought it was doing in 1965, but that it views enforcement as a barrier against immigration rather than as a control mechanism within which intelligent and enhanced immigration can occur. U.S. workers are not helped, U.S. employers are frustrated, and a confused public is not told, and therefore does not realize, why the expansion of immigration, rather than its reduction, is something that they and their families should get behind.

The DOL says it wants to help U.S. workers and this author believes it. They should. That is their mission. Having said that, one wonders if PERM, and the philosophy that is behind it, is the best way to help U.S. workers. Is it possible to help workers without helping the employers who hire them? In the real economy that lives outside of PERM, for example, does such a thing as a "job description" really exist? Do employers have comprehensive job descriptions in mind when they recruit? Do they know precisely what their employees will work on in the future? If there are such job descriptions, is it also true that they never change in any material way for years at a time? Such a static view of the economy-one that few if any advanced nations still embrace, where a command-and-control approach does not allow any deviation, even on grounds of "business necessity"-cannot be reconciled with the way that the U.S. works in the 21st Century.

Most telling is the very notion that there are "normal" requirements for any job today. This is at the heart of PERM. Who is to be the High Priest of Normalcy? By what criteria is normalcy to be determined? Even if you could identify "normal" requirements, is there to be no variation depending on what the employer does, what it can afford, what its strategy is, who its customers are, where the employer is located, what the economic conditions are, and a whole host of similar factors that affect any business decision? Most troubling is PERM’s underlying assumption that workers who satisfy certain bare bones requirements are all the same, fungible and interchangeable parts on an assembly line. Here is the real problem and the real reason why PERM’s approach will stifle the U.S. economy and make it less likely that employers can hire U.S. workers. The notion that human talent and intellectual creativity can be standardized, quantified, and reduced to rigidly objective criteria without any tolerance for individual creativity or imagination is simply flat wrong-so wrong that it would make U.S. business less nimble, less able to respond to unexpected challenges or take advantage of sudden opportunities. Where were Intel, Oracle, or Sun Microsystems 20 years ago? How likely is it that they would have become what they are today if government policies and business decisions considered only what was "normal" at the time and did not allow for new talents or different skill sets that were little noticed at the time but promised a better future that only innovation could deliver?

Does PERM help U.S. workers by not allowing small and medium-size businesses that hire new workers and drive the economy to require knowledge of emerging technologies that have yet to become "normal" but are the key to the future? The question answers itself. Such a policy is a job killer and a guarantee of flat or no growth for years to come.

It did not have to turn out this way. When Congress was debating how to change labor certification in 1965, opponents charged that federal regulators inevitably would go too far. No, said the apostles of change, restraint is possible. Such a debate seems quaint, perhaps even naïve now, but it was not so before the true consequences of building an entire system of labor controls on a negative premise became glaringly apparent. Hope for intelligent change in labor certification was still possible: "Although some fear has been expressed that the Secretary’s power is too broad and too flexible, I am encouraged to believe that, in the face of a continuing congressional oversight, the Secretary will use his power prudently."131 Yet, despite such high hopes, the DOL has proceeded to micromanage U.S. employers far beyond anything contemplated by Congress, as underscored by the PERM rule. Congressional oversight has been manifestly lacking. PERM seeks to regulate what a carpenter can be paid, what safety rules apply to a coal miner or how long a garment worker can toil at her machine.

Such an approach could be applied in the industrial age, but does it apply to the post-industrial knowledge-based economy? Here, in the context of labor certification, the answer is a loud and clear "no." The alien beneficiaries of labor certification have nothing in common save the belief of employers that their talent and creativity makes them worth keeping. Indeed, what distinguishes them is their lack of commonality, their intense individual expertise. If this is so, then any system that, by its very nature, cannot tolerate such subtle distinctions or individual nuances must become divorced from, and a stranger to, the very economic forces that it seeks to shape and control. If it cannot perform this function, there is no way that PERM will ever benefit the very U.S. workers whose interests are its primary justification.

The DOL does a disservice to the U.S. workers it seeks to protect by depriving their employers of the ability to compete in the world economy on which everyone depends. PERM will result in more unemployment, less opportunity, and greater insecurity for all Americans. That is why PERM is not good for the U.S. There must be a better way if we, all of us as one, have but the will to find it. The key is to get the issue back before Congress and litigation is the only way to do that. Knowing how we got to where we are today will not do the U.S. any good unless the courts force Congress to deal with the consequences of the 1965 compromise. Whether that will happen is very much an open question whose answer will go far to determine what kind of a labor certification system the U.S. will have in the years to come.

1 Statement of Michael Murray, Vice President, Resources and Administration, Microsoft, before the Senate Judiciary Committee on Temporary Visa Program for High-Tech Workers (Feb. 25, 1998). 9200009025 Federal Document Clearing House.
2 Labor Certification for the Permanent Employment of Aliens in the United States; Implementation of New System, 67 Fed. Reg. 30466-521 (May 6, 2002) (proposed rule), discussed and reproduced in 79 Interpreter Releases 699 (May 13, 2002).
3 "The dysfunctionality of the so deep that it should lead to major reform of the program." Bell, "DOL’s Audit of Permanent Labor Certification Program: An Analysis," 17 Immigr. Law Rep. at 86 n.8 (1998).
4 See discussion, infra.
5 See discussion, infra.
6 The statutory language is "able, willing, qualified, and available" at INA § 212(a)(5)(A), 8 USCA § 1182(a)(5)(A).
7 See discussion, infra.
8 As a consequence of the Workforce Investment Act of 1998 (WIA), 29 USCA § 2916, which was a major congressionally mandated change in the structure of the Employment and Training Administration (ETA), labor certification is no longer within UI but is now part of an amorphous group called "National Programs" at the DOL. Despite the move to National Programs, however, the labor certification program has not been moved to the Employment Standards Administration, which is the DOL´s enforcement arm. Organizationally, the program is still in the ETA, which is the services wing of the DOL. Even today, years after the WIA, the website for the foreign labor certification program within the DOL is still located at
9 See discussion, infra.
10 See discussion, infra.
11 Papademetriou and Yale-Loehr, Balancing Interests: Rethinking U.S. Selection of Skilled Immigrants, Ch. 3, at 2 (Carnegie Endowment for International Peace, 1996).
12 Papademetriou and Yale-Loehr, supra note 11. In those days, U.S. immigration laws operated under the "national origin quota" system. Many of the potential sources of immigrant labor were in countries with very small quotas. Congress thus opted for the quota as a form of "macro" labor control. The provision was considered largely a standby measure to be used only in dire emergencies. In fact, as Papademetriou and Yale-Loehr observed, the DOL rarely issued negative certifications and did not issue any until 1957.
13 Rodino, "The Impact of Immigration on the American Labor Market," 27 Rutgers L. Rev. 245, 252-53 (1974).
14 Bodin, "Labor Certification Programs," Tenth Annual Immigration and Naturalization Institute Course Handbook 27, 30 (New York: Practising Law Institute, 1979).
15 Id.
16 Gordon, "The Need to Modernize Our Immigration Laws," 13 San Diego L. Rev. 1, 12 (1975).
17 Id.
18 Act of Oct. 3, 1965, Pub. L. No. 89-236, 79 Stat. 911, § 1182(a)(14) (1970). Now INA § 212(a)(5)(A), 8 USCA § 1182(a)(5)(A).
19 "This bill provides that no immigrant will be granted a visa until the Secretary of Labor has made an affirmative finding." 111 Cong. Rec. H21579 (daily ed. Aug. 24, 1965) (statement of Rep. Poff (R-Va.).
20 111 Cong. Rec. S24228 (daily ed. Sept. 17, 1965).
21 111 Cong. Rec. H21586 (daily ed. Aug. 24, 1965).
22 Id. at H21579.
23 Id.
24 North, Alien Workers: A Study of the Labor Certification Program 27 (U.S. Dep’t of Labor, 1971).
25 Nat’l Commission for Manpower Policy, Manpower and Immigration Policies in the United States 53 (1978).
26 Papademetriou and Yale-Loehr, supra note 11, Ch. 3, at 2 n. 4.
27 Pesikoff v. Sec’y of Labor, 501 F.2d 757 (D.C. Cir. 1974), cert. denied, 419 U.S. 1038 (U.S. 1974).
28 111 Cong. Rec. 24227 (daily ed. Sept. 17, 1965) (statement of Sen. Edward M. Kennedy (D-Mass.)).
29 Id.
30 111 Cong. Rec. 24739 (daily ed. Sept. 22, 1965) (statement by Sen. John Tower (R-Tex.)).
31 North, supra note 24 at 28.
32 S. Rep. No. 89-748, at 15 (1965).
33 Id.
34 Id.
35 8 USCA § 1182(a)(5)(A), originally enacted as 8 USCA § 1182(a)(14).
36 111 Cong. Rec. H21572 (daily ed. Aug. 24, 1965).
37 Id. at H21579.
38 Id. at H21586.
39 Id. at H21579 (daily ed. Aug. 24, 1965) (statement of Rep. Ichord).
40 Singhal, "Labor Certification under Revised Regulations," 51 S. Cal. L. Rev. 823, 827 (1978).
41 Id. at 828.
42 5 USCA § 551(6) (1966).
43 5 USCA § 556(d) (1966).
44 8 USCA § 1182(a)(5)(A).
45 Singhal, supra note 40, at 828.
46 Hearings on H.R. 2580 Before the House Subcommittee on Immigration and Naturalization and the Committee on the Judiciary, 89th Cong. 340 (1965) (statement of Andrew Biemiller, AFL-CIO Legislative Director).
47 Id.
48 Hearings on S. 3074 Before the Senate Subcommittee on Immigration and Naturalization and the Committee on the Judiciary, 94th Cong. 59, 60 (1976).
49 Hearings on S. 3074 Before the Senate Subcommittee on Immigration and Naturalization and the Committee on the Judiciary, 94th Cong. 59, 61 (1976) (statement of Ben Burdetsky, Deputy Assistant Secretary for Employment and Training, DOL).
50 Id.
51 Id. at 86.
52 Hearings on S. 3074 Before the Senate Subcommittee on Immigration and Nationality and the Committee on the Judiciary, 94th Cong. 143, 151 (1965) (statement of Andrew J. Biemiller, AFL-CIO, as presented by Kenneth Meiklejohn).
53 Bodin, supra note 14, at 37.
54 Id. at 35–36.
55 30 Fed. Reg. 14494 (Nov.19, 1965).
56 30 Fed. Reg. 14979 (Dec. 3, 1965).
57 31 Fed. Reg. 16412 (Dec. 22, 1966).
58 32 Fed. Reg. 867 (Jan. 23, 1967).
59 36 Fed. Reg. 2462, 2463 (Feb. 4, 1971).
60 29 CFR § 60.1 (1975).
61 “Willing” is an attribute of an individual, not of a statistical agglomeration of allegedly “willing” U.S. workers.
62 See legislative history discussion, supra.
63 42 Fed. Reg. 3440 (Jan. 18, 1977).
64 Gordon, supra note 16.
65 Wasserman, “The Labor Certification Program, Its Deficiencies and Appellate Review,” 49 Interpreter Releases 165 (June 15, 1972).
66 Wasserman, “The Labor Certification Program, Its Deficiencies and Appellate Review,” Presentation at 1972 Conference of Association of Immigration and Nationality Lawyers, New Orleans (May 26, 1972).
67 Pesikoff v. Sec’y of Labor, 501 F.2d 757, 771 (D.C. Cir. 1974) (MacKinnon, J., dissenting).
> 68 Singhal, supra note 40, at 843.
69 52 Fed. Reg. 11218 (Apr. 8, 1987). The BALCA suffers from a strange defect: unlike the INS and the BIA where regulations exist that make BIA decisions binding on all officers and employees of the Service and Immigration Judges, BALCA decisions are not binding for DOL and SESA officials.
70 Matter of Artdesign Inc., 89–INA–99 (Dec. 5, 1989).
71 Matter of Delitizer Corp. of Newton, 88–INA–482 (May 9, 1990), discussed in 67 Interpreter Releases 1235 (Nov. 5, 1990).
72 Matter of Francis Kellogg, 94–INA–465 (Feb. 2, 1998), discussed in 75 Interpreter Releases 1449 (Oct. 19, 1998).
73 Editors’ note: GAL 1–97 was re-issued as GAL 1–97, Change 1, and was published in 64 Fed. Reg. 23983–86 (May 4, 1999). That notice, and a discussion of the modified GAL 1–97, are discussed in 76 Interpreter Releases 759 (May 17, 1999). Memoranda clarifying GAL 1–97 were issued subsequently by Dale M. Ziegler, Chief of the Division of Foreign Labor Certification, on March 18, 2002, and May 28, 2002. For more, see Endelman, “Through the Looking Glass: The Impact of GAL 1–97 and the Future of Labor Certification,” 74 Interpreter Releases 1461 (Sept. 29, 1997); Stanton, “Changing Paradigms for Labor Certification: New Opportunities, Strategies, and Risks,” 97–9 Immigr. Briefings 1 (Sept. 1997); Endelman and Osuna, “Labor Certification at a Crossroads: How GAL 1–97 Has Changed Labor Certification Policy and Practice,” 97–8 Immigr. Briefings 1 (Aug. 1997).
74 69 Fed. Reg. 43716–18 (July 21, 2004), reported on and reproduced in 81 Interpreter Releases 969 (July 26, 2004).
75 Letter from Rebecca Gamez, Alaska Employment Security Director, to Mike Brauser, Regional Executive, ETA, U.S. DOL, Seattle, Washington (July 10, 1998), available online at
76 Letter from the SWA Administrators to Mike Brauser, Regional Executive, ETA, U.S. DOL, Seattle, Washington (July 21, 1998), available online at (Signed by each of the Region X State Workforce Agency administrators: Rebecca Gamez, Alaska; Virlena Crosley, Oregon; Roger Madsen, Idaho; Carver Gayton, Washington.)
77 Statement of Joseph Weisenburger, Deputy Commissioner, New Hampshire Dep’t of Employment Security, on behalf of the Interstate Conference of Employment Security Agencies before the House Appropriations Labor, Health and Human Services, and Education Subcommittee on FY 1999 Labor-HHS Approps., Jan. 29, 1998, Federal Document Clearing House 9200006672.
78 H.R. Rep. No. 101-723(I) at 41, reprinted in USCCAN 6721.
78 H.R. Rep. No. 101–723(I) at 41, reprinted in USCCAN 6721.
79 H.R. Rep. No. 101–723(I) at 43, reprinted in USCCAN 6722.
80 Endelman and Walker, “Legislative History Concerning Labor Certification Changes (or Lack Thereof) in H–1B Attestation Requirements of the Immigration Act of 1990,
Understanding the Immigration Act of 1990 at 95, American Immigration Lawyers Association.
81 “Attestations are deemed certified at end of 30-day period after filing unless challenged.” Id. at 101.
82 H.R. Rep. No. 101–723(I) at 61, reprinted in USCCAN 6722, 6741.
83 Originally 8 USCA § 1182(a)(14), recodified as 8 USCA § 1182(a)(5)(A) (1990).
84 H.R. Rep. No. 101–723(II) at 139–140, reprinted in USCCAN 6722, 6777.
85 Applications for occupations listed in Schedule A do not have to go through recruitment, and are thus spared the most onerous and time-consuming part of the labor certification process.
86 Reengineering of Permanent Labor Certification Program; Solicitation of Comments, 60 Fed. Reg. 36440 (July 17, 1995), reported on and reproduced in 72 Interpreter Releases 976, 993 (July 24, 1995)..
87 National Performance Review, Ch. 2, “Getting Results” (Sept. 7, 1995).
88 National Performance Review, “Appendix C: New Recommendations by Agency” (Sept. 7, 1995).
89 Anonymous message posted on Newsgroups:, misc.immigration.usa on 1998/02/21, Message-ID “34EF189E.EE8726E8@a.b.c.nowhere,” originally retrieved from DejaNews [grammatical errors in original].
90 U.S. Department of Labor, Office of Inspector General, Office of Audit, Final Report: “The Department of Labor’s Foreign Labor Certification Programs: The System Is Broken and Needs To Be Fixed,” Rep. No. 06–96–002–03–321, at Appendix II—Agency Response, para. 7 (May 22, 1996), by Timothy M. Barnicle, Assistant Secretary of the ETA.
91 U.S. Department of Labor, Office of Inspector General, Office of Audit, Final Report: “The Department of Labor’s Foreign Labor Certification Programs: The System Is Broken and Needs To Be Fixed,” Rep. No. 06–96–002–03–321 (May 22, 1996).
92 Bell, supra note 3, at 85.
93 S. Rep. No. 105–186 at 22–23 (1998).
94 Bell, supra note 3.
95 Bell, supra note 3.
96 Permanent Alien Labor Certification Program Re-engineering Questionnaire, 60 Fed. Reg. 36444 (July 17, 1995). See 72 Interpreter Releases 976 (July 24, 1995).
97 “One point should be clarified. Your auditors focused on approved cases—since approximately one-quarter of the applications State Employment Security Agencies (SESAs) receive are dropped by employers before any recruitment, approved cases represent about 49% of the total number of applications filed. Of the applications subsequently processed by regional offices, approximately 65% are approved, indicating a return and denial rate of about 35%.) This overall approval rate indicates that the program preserves jobs for U.S. workers to some modest extent. Clearly, strengthening the program would greatly improve protection of U.S. jobs.” Barnicle, supra note 90.
98 Bell, supra note 3.
99 INA § 212(a)(5)(A), 8 USCA § 1182(a)(5)(A).
100 Supra note 90 at Audit Results, paras. 5 and 6 (May 22, 1996).
101 H.R. Rep. No. 101–723(I) at 43, reprinted in USCCAN 6722, 6723.
102 Cong. Rec. S3287 (daily ed. Apr. 15, 1996).
103 Bell, supra note 3.
104 Supra note 90 at Appendix I—Background, The Permanent Program, para. 1.
105 Pesikoff v. Sec’y of Labor, 501 F.2d 757 (D.C. Cir. 1974), cert. denied, 419 U.S. 1038 (U.S. 1974).
106 Pesikoff v. Sec’y of Labor, 501 F.2d 757, 764 (D.C. Cir. 1974).
107 Id. at 768.(Mackinnon, J. dissenting).
108 Id. at 771 (Mackinnon, J. dissenting).
109 Id. at 763, n.9.
110 Silva v. Sec’y of Labor, 518 F. 2d 301, 309 (5th Cir. 1975).
111 Id.
112 Ratnayake v. Mack, 489 F.2d 1207, 1212 (8th Cir. 1974).
113 Digilab v. Secr’y of Labor, 495 F. 2d 323, 326 (1st Cir. 1974).
114 Sherwin-Williams Co. v. Reg’l Manpower Adm’r, 439 F. Supp. 272, 274 (N.D.Ill. 1976). See also Shuk Yee Chan v. R.M.A., 521 F.2d 592, 595–96 (7th Cir. 1975).
115 Sherwin-Williams Co. v. Reg’l Manpower Adm’r, 439 F. Supp. 272, 275 (N.D.Ill. 1976).
116 Bitang v. Reg’l Manpower Adm’r, 351 F. Supp. 1342, 1345 ( N.D. Ill. 1972).
117 Kenall Mfg Co. v. Employment and Training Administration, 1981 U.S. Dist. LEXIS 14960 (N.D. Ill. 1981).
118 Prod. Tool Corp. v. Employment and Training Administration, 688 F.2d 1161, 1170 (7th Cir. 1982).
119 Industrial Holographics Inc v. Donovan, 722 F.2d 1362, n.17 (7th Cir. 1983).
120 Warmtex Enterprises v. Martin, 953 F.2d 1133 (9th Cir. 1992).
121 See De Jesus Ramirez v. Reich, 156 F.3d 1273 (D.C. Cir. 1998).
122 Statement on Employment-Based Immigration Reform by Institute of Electrical and Electronic Engineers (IEEE) before the Senate Subcommittee on Immigration, Oct. 13, 1995.
123 Supra note 90 at Audit Results, paras. 5, 6, 7, and 8.
124 Barnicle, supra note 90 at Appendix II, paras. 2 and 3.
125 Bell, supra note 3.
126 Bell, supra note 3.
127 67 Fed. Reg. 30466-521, 30468 (May 6, 2002), discussed and reproduced in 79 Interpreter Releases 699 (May 13, 2002).
128 8 USCA § 1182(a)(5)(A).
129 8 USCA § 1361.
130 Mantle, "PERM Labor Certification: Justifying the Proposed Regulations," available online at,0617-mantle.pdf and,0618-mantle.pdf.
131 Cong. Rec. H21579 (daily ed. Aug. 24, 1965) (statement of Rep. Poff (R-Va.)).

About The Author

Gary Endelman practices immigration law at BP America Inc. The opinions expressed in this column are purely personal and do not represent the views or beliefs of BP America Inc. in any way.

** The author extends special thanks to D. Ray Mantle of Brigham Young University for his extraordinarily hard work and invaluable research assistance. Courtney Rineer helped in polishing the final manuscript of this article. The author also is grateful to Mark Mancini for making available the case file that the great litigator Jack Wasserman compiled on the Pesikoff case. Finally, the author extends special thanks to Sam Udani, "whose friendship and intellectual partnership over the past seven years have transformed my own thinking on labor certification and patiently brought me to the point where I could write this article. The good ideas are Sam´s; whatever inadequacy of thought and expression may remain are my own."

The opinions expressed in this article do not necessarily reflect the opinion of ILW.COM.