June 26, 2002
Mr. Chairman and Members of the Committee: I appreciate the invitation to appear before you to discuss a vitally important topic. Serious planning for reorganization of the government's immigration management function was well under way at the end of 1997 when I left office as General Counsel of the Immigration and Naturalization Service (INS) to return to teaching. It is most unfortunate that we are still debating and not implementing, after nearly five years of discussion, and despite wide agreement on the basic contours of the needed reorganization. The lingering uncertainty has damaged INS staff morale, caused some highly capable people to leave the agency, and hampered other sorts of reforms. Now, of course, President Bush's proposal for a new Homeland Security Department has introduced an additional complication into the discussion -- just as it appeared that we were on the threshold of enacting a law that would have permitted reorganization within the Department of Justice (DOJ) to begin in earnest. Despite this, I hope the Congress can settle the basic structural decisions soon so that we can move on into the implementation phase.
I. The Limits of Reorganization Remedies
Reorganization has almost always been oversold as a solution to operational, management, and policy problems. Professor Harold Seidman once observed that these kinds of restructuring initiatives are the "twentieth century equivalent of the medieval search for the philosopher's stone. . . . If only we can find the right formula for coordination, we can reconcile the irreconcilable, harmonize competing and wholly divergent interests, overcome irrationalities in our government structures and make hard policy choices to which no one will dissent."
His characterization of inflated and unrealistic expectations applies fully to much of the rhetoric that has surrounded INS reorganization, and I fear it will prove the same with the new Department of Homeland Security. It seems that we often veer off into debates over organization when we cannot quite bring ourselves to face difficult policy choices squarely or cannot achieve the clarity of mission or resource commitments that are usually far more important. Whatever may be true of organization for homeland security generally, I am convinced that the most important fixes for many of the problems of our immigration system will stem not from reorganization but from smaller scale and less showy changes. Real improvements must derive from patient, detailed work that addresses specific functions one by one, identifies weaknesses, thinks hard about competing solutions, selects one, and maps out a long-term plan to make sure the change is well-implemented on the ground. Too often, reorganization talk becomes a way of changing the public focus and postponing difficult decisions. At the very least, reorganization, especially on the scale contemplated for the Homeland Security Department, will divert energies and postpone this more productive micro-focus on specific improvements.
The timing of the current Homeland Security proposal, which would move the immigration function to a new and untried department, is at least ironic. Immigration was long a neglected stepchild within the Department of Justice, but that situation had been turned around over the last decade. Attorneys General were getting the hang of paying close attention to immigration, of understanding its importance not only to law enforcement but also to America's flourishing -- to making us the diverse, vibrant community we are and want to remain. I fear that immigration will become a stepchild anew in the new department -- although I offer recommendations later that may help avoid that outcome.
II. Guiding Principles
Based on my twenty-odd years of teaching and writing about immigration, interspersed with government service in the Departments of State and Justice, I suggest here a few key points that should be kept in mind as the Congress hammers out the details for implementing the reorganization of the immigration function. I then turn to a few more specific recommendations.
1. Immigration is about more than just enforcement and dangers. The attacks of September 11 reminded us all about the shortcomings of our immigration enforcement system and the dangers that some migrants can pose for our country. That is a vital lesson, and I hope it will help invigorate sensible, resolute immigration enforcement -- not only those elements of enforcement directed at national security dangers, but also the more routine enforcement against those who have not honored our laws and have no legitimate defense or excuse for their actions. Widespread violations, even though most of the violators are hard-working and benign, erode respect for the rule of law. They create a climate, including the widespread use of false identification, that makes it harder to protect against that small minority of illegal migrants who are involved in criminal or terrorist actions.
But having said that, I want to emphasize that immigration policy is not just about guarding against dangers and violations. We are truly a land of immigrants, and this country has benefited enormously, enriching our arts, our culture, our cuisine, and our economy in the process. Our immigration policy also makes a vital contribution to the protection of the world's refugees. Indeed, it is a singular shame that the one slice of permanent immigration most sharply affected by September 11 appears to be refugee admissions -- suspended for several months and not yet back up to the needed pace of entries. Whatever reorganization plan the nation adopts, we have to restore our historic position as a leader in refugee resettlement. Moreover, we must not let the fears unleashed by the September attacks diminish the human bridges to other nations that immigration provides -- through those who come both for temporary visits and for permanent resettlement. Immigration policy serves to reunite families, to enrich our schools and colleges, to educate visitors on the realities of American life (thereby puncturing distorted pictures that our enemies spread), to educate Americans about the rest of the world, to win us friends in other countries and their governments, and to serve our historic humanitarian objectives.
Losing sight of the positive side of immigration, of its richness and value, may be the greatest risk posed by placing the immigration function in a new Department named Homeland Security.
2. Keep immigration services and enforcement closely linked. Most INS reform proposals have focused on dividing immigration enforcement (carried out by such units as the Border Patrol, investigations, detention and removal, and intelligence) from immigration services (the approval of applications and petitions filed by would-be migrants or citizens or their sponsors -- sometimes called adjudications or benefits). Indeed there has been widespread agreement for several years on some such split as the central component of INS reorganization plans. Both the Sensenbrenner-Conyers bill (H.R. 3231) and the Kennedy-Brownback bill (S. 2444) employ this framework, as did the Bush Administration's initial INS restructuring plan. Many have supported this scheme as a way of assuring that the services component is given adequate attention and resources. And reorganizing INS along these lines held modest promise for relocating field offices and consolidating lines of authority in ways that could help change the agency culture and reduce the risk that district offices act like separate fiefdoms with their own divergent approaches.
But as the INS reorganization debate has proceeded, more and more people have come to realize the potential disadvantages of a too-rigid split between enforcement and adjudication. Before the President's announcement on June 6 of his new Homeland Security reorganization proposal, much of the most recent commentary focused on the need to assure that the two components are efficiently linked and always work closely together. The Kennedy-Brownback bill, in my view, would do a better job than the House-passed bill in honoring this imperative. It provides a better staffing arrangement for the director of its new Immigration Affairs Agency (including a single General Counsel's office) than is specified for the comparable official (an Associate Attorney General) under H.R. 3231. It also provides a wider range of authority for the director, including clearer chains of command, thereby creating a far better institutional home for functions that inevitably must be shared or closely coordinated between enforcement and services.
Unfortunately, the prospective move of immigration functions to Homeland Security has revived some of the old discussion about a more complete split -- perhaps leaving immigration services in the Justice Department and moving immigration enforcement (or some segment of it, such as the Border Patrol) to the new Department. I understand the impulse behind that thought. As indicated under principle 1, there is a genuine risk that the services side of immigration management will be diminished as a result of the Homeland Security reorganization. The theory would be that leaving services in a different department could help avoid that outcome.
In my view, however, that theory would not be realized in practice. Such a split would almost surely leave services even weaker than placing them in Homeland Security. Moreover, it would bring in its train other considerable disadvantages by splitting up what really should be considered a single package of immigration management, encompassing both services and enforcement. The immigration function should remain unified, but a few revisions to the President's plan can better honor the positive features of immigration and sustain recognition of the importance of adjudications.
Why not a more complete split? Services obviously should not be provided to persons who are disqualified; very often the relevant information concerning the disqualification will have been developed through immigration enforcement efforts. Although it is certainly possible to develop cooperation and information-sharing arrangements across agencies and with diverse departments (INS has done so for decades with the Departments of State and Labor, and with other components of the Department of Justice), such cooperation works better when it takes place within the same agency. Most significantly, we are now in a phase of our national life where all immigration benefits will be conditioned on some form of screening against enforcement-related information -- whether a name check against a lookout database or a demand for fingerprints or a detailed interview. To leave immigration services as a vestigial unit in the Justice Department (or perhaps some other department besides Homeland Security) is to make it hostage to decisions on funding and priorities made primarily in other venues. Timely provision of such information to immigration adjudicators could easily suffer, and securing a remedy for the delays might well require mobilizing one Cabinet-level official to take up the matter with another such official. It seems unlikely that the services unit, if left standing solo, will be able to place a high-priority claim on the time of a Cabinet officer. Immigration services are more likely to have the bureaucratic clout they need if they remain as part of an overall immigration function kept together in the same department.
I also fear that a wholly separated immigration services bureau would fall under a constant cloud of suspicion that it was insufficiently attentive to security concerns in its decisions. Such a reaction could affect both funding and ongoing congressional and public support. Although I hope we can develop a system where adjudications are entirely funded from fees at a level that assures timely decisionmaking, I suspect that additional funding will often be necessary. If so, services are more likely to claim that kind of support and attention if they remain in the same department with immigration enforcement.
For all these reasons, if the immigration function, as a whole, cannot remain in Justice (some move now appears overwhelmingly likely), then it is far better to transfer all functions together to the new Department, provided that the Department itself is properly structured to counter the real risk that immigration services will be neglected.
3. Immigration deserves ongoing attention from the highest levels of government. As I mentioned earlier, the last decade has seen considerable gains in the salience of immigration policy. On the positive side, immigration is a vitally important part of our national tradition and a key to many of our country's successes. On the negative side, neglecting immigration functions leaves us vulnerable to intense dangers from terrorists as well as a more subtle erosion of sound governance and respect for the rule of law. Moreover, immigration, the movement of people, is sensitive in a way that movement of cargo will never be. Family hopes and expectations rest on immigration; immigration policy can either facilitate or thwart the development of effective business plans in a globalized economy; and we must never forget that immigration policy also makes way for vital humanitarian initiatives, primarily through refugee resettlement. Regulation, management, and sometimes facilitation of human migration therefore calls for a very different mix of skills, techniques, and orientation than are needed in the other units projected for the Border and Transportation Security Division of the Homeland Security Department.
Further, we need to structure any agency with immigration responsibilities so that immigration issues receive attention from the highest levels of government whenever the need arises. The bill recently passed by the House recognized this level of priority by elevating the head of the immigration function to the rank of Associate Attorney General, meant to be one of the top four or five officers in one of the most powerful of Cabinet departments. The Homeland Security proposal of the President would reverse this useful trend. Immigration would become just one of six components in the Border and Transportation Security Division. Presumably the head or heads of immigration would report to an undersecretary, who would in turn report to a Cabinet secretary. That placement is a step in the wrong direction in a new century when immigration issues are likely only to increase in importance.
III. Recommended Solution: an Undersecretary for Immigration Affairs in the Department of Homeland Security
Given the momentum generated by the proposal to create a Department of Homeland Security, the best way to serve all the values identified above is to keep the full immigration function in that Department but to alter the structure. In my view, the immigration function should be placed in a separate, fifth division of the Department, under an undersecretary for immigration affairs. Immigration is sufficiently important, both to enforcement and to the human values and cultural and economic successes that characterize this land of immigration, that it merits such a structure.
An undersecretary who can focus solely on immigration is far more likely to keep in mind the full richness of this area of public policy -- and hence to assure sufficient priority for the services side of the house -- than is an undersecretary who must also focus on the demands, largely enforcement-related, of five other components. When particular needs arise in the immigration services realm, they are more likely to be addressed by an immigration undersecretary -- and to be raised to the level of the secretary if needed -- than would be the case with a residual services unit left behind in Justice. Moreover, this official, with a range of authorities similar to those given the director under the Kennedy-Brownback bill, can craft and oversee ongoing coordination between enforcement and services, and can assure the public that immigration benefits are not being extended without adequate screening and appropriate attention to security risks.
Such a restructuring of the President's proposal would not undercut the principal benefits claimed for the original Homeland Security plan. Although the new Department would have five divisions instead of the original four, this separation is justified by the unique challenges that arise in the management of the movement of human beings. Nearly all of the benefits to coordination that were expected from moving immigration, along with Customs, Transportation Security, the Animal and Plant Inspection Service, the Federal Protection Service, and the Coast Guard into the same division can still be realized, because all of these units remain within the same Homeland Security Department.
A fifth division would also bring better balance among the divisions of the new Department. Under the Bush plan, the Border and Transportation Security Division (including immigration) would be a behemoth, involving 156,000 employees (over 90 percent of the department staff) operating under current budgets totaling nearly $24 billion. The other three components combined would have 13,000 employees and a budget of $14 billion. Splitting off immigration would create a fifth division with a budget of at least the current $6.4 billion, with staffing of over 39,000 -- plenty to justify a separate unit. But size itself is not the cornerstone of the case for a separate Immigration Affairs Division. Instead such a change primarily serves to recognize the unique challenges posed by and the unique values that derive from immigration policy.
Considering a different alternative. I have already indicated why I consider this arrangement superior to a plan that would keep immigration services in Justice while moving all immigration enforcement to Homeland Security. But I should say a word about its superiority to another alternative that is reflected in the Hart-Rudman Commission's recommendations and has found its way into some recent legislative proposals. That is the idea of moving only the Border Patrol, but not the rest of immigration enforcement (nor services), to Homeland Security.
Such a move would be a mistake. The Border Patrol's work is more closely linked with the rest of immigration enforcement (as well as other immigration functions) than the Hart-Rudman proposal acknowledged, and to split it off alone would bring far greater obstacles to effective coordination than any gains one might expect from placing the Patrol in the same Department as the Coast Guard and Customs. Take an example that demonstrates these linkages. As the Border Patrol's effectiveness along the southwest border increased throughout the 1990s through the increased use of forward deployment (including Operations Hold the Line and Gatekeeper), we saw a real change in the kind of challenges faced in the ports of entry, staffed not by the Border Patrol but by the Inspections division of INS. The new Border Patrol strategy thwarted smugglers who formerly guided people across relatively unpoliced stretches of the border. But the smugglers did not simply go out of business. They responded to the new strategy by promoting or facilitating fraudulent claims in the ports of entry, and through more sophisticated concealment and transportation schemes for those who still made it through.
Smugglers (and no doubt terrorists) continually respond to enforcement changes through new strategies of their own, which must be met, in turn, by new moves from the US government. All parts of the immigration management system, and not simply the Border Patrol, must be mobilized in this process. The response must include classic service or adjudication functions, which may find themselves challenged with more sophisticated fraud or manipulation, and must certainly include the inspection function, which partakes of both adjudication and enforcement. Border enforcement, in short, is intimately linked with interior enforcement as well as with immigration services. Its departmental placement should reflect that reality. The Border Patrol should be kept together with all other elements of the immigration function.
IV. Further Issues
The visa function. It is not clear from the accounts I have seen or from the skeletal legislative package just what is intended for the visa screening and issuance function under the Bush Administration's proposal. If we are going to undertake major restructuring of immigration functions, we have to examine all components of the process. The issuance and denial of visas really constitute the first line of border protection. Although cooperation between INS and State's Consular Affairs Bureau has worked reasonably well, there is a strong case to be made to place this function together with other immigration components in Homeland Security, while still recognizing some of the unique needs and competencies involved. The case for such a transfer is particularly strong if Congress alters the package, as I have recommended, to create a separate Immigration Affairs Division headed by an undersecretary.
EOIR. The Executive Office for Immigration Review consists of the Board of Immigration Appeals (BIA), the immigration judges, and a small number of administrative law judges (who deal with specialized immigration-related cases, mainly employer sanctions). EOIR handles highly formalized adjudications with high stakes, notably removal cases and related questions (such as bond redetermination). Where this component should be located poses some special challenges.
I disagree with those who call for EOIR to be made a separate and independent adjudicative agency. Although it is vital that the judges and Board members retain complete insulation and independence in their decisions on individual cases, a sound immigration control system requires that they be managerially linked with the rest of the immigration function. For example, sometimes immigration judges need to be deployed on a priority basis, along with other immigration management resources, to an area experiencing a sudden migration influx. Such an episode occurred in South Texas in 1988, and the speedy creation of additional court capacity there helped resolve many deportation cases quickly and thereby sent a needed deterrent message. Contingency planning exercises in which I participated during my tenure as INS General Counsel in the 1990s always included potential arrangements for deployment of immigration judges to the temporary facilities, perhaps operated in tents or other hastily erected shelters. Although it is not impossible to do such planning or deployment across department or agency lines, deployment is much more likely to be successful -- and swift -- if such managerial and logistical decisions can be made by the same official at the head of both chains of command. And in the immigration business, speedy reaction may prove essential in keeping a minor situation from turning into a major crisis. That kind of managerial authority at the cabinet or undersecretary level can be provided in a way that is wholly consistent with independence in deciding individual cases.
I recommend transferring EOIR, like other immigration components, to the Department of Homeland Security. There the director of EOIR should report to the Undersecretary for Immigration Affairs, but individual immigration judges and BIA members would retain all current guarantees of independence in making decisions in individual cases. Congress might also profitably consider additional measures to bolster that independence, but in a framework that recognizes the managerial and logistical linkages.
"Referral" of BIA decisions. Although this issue is a difficult one, it probably makes sense for the Secretary or Undersecretary of Homeland Security to have the authority to hear what amount to appeals or plenary rehearings from the decisions of the BIA, under the infrequently invoked "referral" authority that now appears in 8 C.F.R. § 3.1(h). Under that regulation, BIA cases are now referred to the Attorney General in three circumstances: when requested by the INS Commissioner, when requested by the BIA majority or its chairman, or upon the Attorney General's own direction. Sparingly used (an average of once or twice a year), this procedure has historically been a useful device to assure development of the law in accord with the Attorney General's broad responsibilities in the field of immigration. The immigration laws, detailed as they are, inevitably leave room for interpretive and policy discretion. Because immigration management is often closely connected with foreign policy and other broad vistas of national policy, as September 11 reminded us, the final responsibility for exercising that discretion should rest with a high-level and publicly accountable official who sees more of the policy horizon than does the BIA. Referral to the Attorney General has provided such a mechanism. At the same time, the AG's exercise of that carefully structured authority has always been subject to judicial review -- an important safeguard to assure that the decision truly remains within the bounds of discretion set by statutory and constitutional law.
Such a procedure should not be seen as interference with the decisional independence of the Board -- any more than an authoritative ruling by the Supreme Court is considered to impair the independence of trial judges. The judges must henceforth follow the doctrine propounded in the precedent decision, but they are certainly under no obligation to take phone calls from the Chief Justice directing the outcome in individual cases. Referral similarly follows a highly structured and formalized path that preserves the insulation of the Board and the immigration judges and disciplines the intervention of the Attorney General. It results in a detailed, public, quasi-judicial decision, with a full statement of reasons justifying the result and giving guidance for future cases. And as indicated, it remains subject to judicial correction.
I therefore believe that referral should be retained, but it may require some special arrangements. If EOIR is moved from Justice to Homeland Security (as is preferable if the rest of the immigration function is transferred), then referral should no longer go to the AG but should instead go to the equivalent official in Homeland Security. Whether that would be the Cabinet Secretary or the Undersecretary is probably less important than assuring high-level, high-competence legal advice and support when a referral occurs. That legal advice should not come from the General Counsel's office that normally handles the legal business of the enforcement and services bureaus, nor probably from the comparable office at EOIR. It might make sense to retain the services of the Office of Legal Counsel in the Department of Justice (which has historically played this limited advisory and support role in referrals to the AG), but perhaps the office of counsel to the Secretary of Homeland Security could perform this specialized and infrequent function instead.
V. Final Recommendations
Reorganization is upon us, and it is vital to structure it in a way that is true to the unique complexities and unique values inherent in immigration policy. I close with two final thoughts that are meant to help keep reorganization in perspective.
1. We need fewer studies and audits, and more capacity to implement the reports generated by past studies of flaws in our immigration system. Numerous studies, by the General Accounting Office (GAO), the DOJ Inspector General (IG), and by outside consultants and scholars, have documented weaknesses in our immigration management system and mapped out suggested fixes. The problem is not a lack of such reports; it is instead their hyperabundance.
Each suggested fix, no matter how straightforward it appears in the clinical prose of the auditors, in fact requires enormous energy and focus to implement. For one thing, the suggested reform is always contestable; sometimes outside reviewers overlook certain key operational elements well known to those who carry out immigration tasks in the field day in and day out. Different schools of thought have to be heard out; a decision made by responsible managers, often (until now) the Commissioner or the Attorney General, who obviously have numerous competing demands on their time; and then new regulations, forms and practices have to be put in place, often including thousands of hours of retraining. Regulations alone can easily get stalled for months or years in the clearance process unless a high-level official personally keeps the heat on. No commissioner, director, or undersecretary can be successful at this kind of reform if spread too thin. Choices have to be made about which reforms to press first, so that they can be seen through to actual realization on the ground. Other reforms will simply have to wait.
This reality of immigration reform is often lost from view in oversight hearings that criticize immigration officials for failing to implement some GAO or IG recommendation. There are simply too many of them, measured against the reality of what it takes to turn such proposed reforms into reality. Sometimes during my tenure at INS, while scheduling yet one more interview with the IG's extensive staff examining some immigration misfire, my thoughts wandered to how nice it would be if INS could simply incorporate the investigative or audit team to augment the meager ranks of those actually trying to implement needed changes. My point is that we must take care lest audits, investigations, and reports claim a disproportionate share of the time and resources that could instead be used in implementation. We also need to be realistic about the need to prioritize those smaller scale, but vital, fixes to our immigration machinery.
In this connection, I would advise against creating an ombudsman in the new department, although I well understand the frustrations that have led to the popularity of such a proposal in nearly all the current immigration reform bills. Instead of adding one more wild card in the chain of authority and command, it would be wiser to concentrate on building better customer service, with enhanced quality control, directly into the functioning of the new immigration services bureau. If an ombudsman is still thought necessary for individual cases, I would urge keeping the office as lean as possible, thus freeing up resources for improvements in direct services. Above all, the legislation should greatly limit the responsibility and authority of the ombudsman to make broad public reports and detailed suggestions for management reforms. These functions are already adequately served by the GAO and the IG's office.
2. Clarity of mission is more important than reorganization to the eventual success of immigration policy. Congress should actively work to simplify the immigration laws. Closely related to the previous point is the overwhelming complexity of our modern immigration laws. INS has had little time to work on patient micro-level improvements over the last six years, because it has been so overwhelmed by the need to implement, often on a compressed deadline, a variety of complex new statutory initiatives. If we want successful immigration management, we have to provide a stable structure of laws, and we should seek to simplify them wherever possible. Instead, we are going in the opposite direction. Much of the problem derived from the 1996 reform legislation, particularly the Illegal Immigration Reform and Immigrant Responsibility Act, which was both massive and unnecessarily complex.
Even where Congress has recognized that some of those provisions were ill-designed or overly harsh, it has not taken the better step of repealing the problematic provision. Instead it seems bent on adopting complex new carve-outs that help only a fraction of those in question. This happened, for example, with the enactment in 1996 of tight new restrictions on what we used to call suspension of deportation (it is now cancellation of removal under INA § 240A(b)). Congress seemed to regret that step in 1997, but instead of restoring the old rules, it adopted a special set of exemptions (a kind of amnesty) applicable only to Nicaraguans and Cubans, plus yet a different set of rules, less generous, for certain Guatemalans and Salvadorans. (The Nicaraguan Adjustment and Central American Relief Act.) Creating the new regulations and forms consumed an inordinate amount of INS time and effort that could have been better spent elsewhere. The problem was compounded the next year, when Haitians were (in my view properly) brought into the fold of exemption -- but with yet another set of special rules not quite like those that apply to the earlier groups. (The Haitian Refugee Immigration Fairness Act of 1998.)
The debate over section 245(i) of the Immigration and Nationality Act (INA) unfortunately follows the same pattern. In all the debate over this provision, one can hardly tell that the meat of the problem derived from a superficially attractive, but ultimately quite flawed, enforcement measure adopted in 1996. That measure consists of the three- and ten-year bars of INA § 212(a)(9)(B), meant to punish people for earlier illegal presence in the United States. At various points Congress has obviously grown uncomfortable with full enforcement of the bars, but instead of repealing them, it has adopted a highly complex set of grandfathering provisions, through periodic revivals of 245(i), meant to free some persons from the bars (but only doing so imperfectly). We could hardly have designed a response to the underlying problem more certain to complicate INS's implementation task. Repeal of the bars would be a far better course, particularly because of its beneficial impact in freeing up INS resources for more productive use.
Legal complexity is not necessarily bad. Sometimes it is needed to address genuinely complex problems. But far too much of the INA consists of complications that cannot claim such a justification -- complexity that solved a transitory political problem but only at the unobserved cost of disabling sound immigration functioning and diverting reform energies.
The Immigration and Nationality Act turns 50 years old tomorrow. To my knowledge, no one is celebrating the anniversary, in part because it has become such an unwieldy and hard-to-manage set of rules and procedures. For the future, Congress should adopt no new immigration measures without considering carefully their full implementation impact. Whenever a problem could be solved by a simple repeal, that course should be chosen over any new mechanism that adds special benefits for a targeted group. We have driven ourselves into many of our current immigration problems because we have ignored the realities of implementation when new legislation is proposed and debated. We should at least stop compounding the difficulties. Maybe at some point, once reorganization is well under way, we could embark on a deliberate course of immigration law simplification.