William Holden
Chief, Driver Register
Docket Management
Room PL-401
National Highway Traffic Safety Administration
Nassif Building
400 Seventh Street, SW
Washington, D.C. 20590

August 3, 1998

Dear Mr. Holden:

Thank you for the opportunity to comment on Docket number NHTSA 98-3945. The National Conference of State Legislatures (NCSL) appreciates the opportunity to discuss the regulations to implement the requirements contained in section 656 (b) of the Illegal Immigration Reform and Immigrant Responsibility Act of 1996, entitled State Issued Driver's Licenses and Comparable Identification Documents.

The issuance and production of driver's licenses has always been within the domain of the states. Currently, the issuance of driver's licenses is within the purview of the states. NCSL believes that the authority to issue driver's licenses should remain with the states.

After careful inspection of this Notice of Proposed Rulemaking by state legislatures and staff across the country and NCSL staff, we have profound concerns about many aspects of the both the federal law and the NPRM. These concerns are:

1. The preemption of state authority resulting from the enactment of the law and the provisions contained in the rule
2. The movement by the federal government to include social security numbers (SSN) on driver's licenses despite the fact that many states specifically prohibit the inclusion of SSNs on licenses.
3. Contrary to the language in the Unfunded Mandate Reform Act of 1994 (UMRA)(P.L. 104-4), no state legislators, other elected officials, or their national organizations were consulted to "provide timely and meaningful input" in the development of this regulatory proposal.
4. The costs estimated by the DOT to correspond to the requirements in the NPRM are grossly underestimated. According to information submitted from the states, the cost to implement these directives would well exceed $100 million, resulting in a direct violation of the UMRA. Additionally, it appears that not only are the costs provided by the DOT flawed, but also the process used to obtain the data is questionable.

Additionally, it is quite apparent that there is a larger cost associated with producing a license with a holographic or "ghost" image than producing one without. The cost to change over from the production of the state's current license process to one with larger overall costs only adds to the already excessive financial burden imposed on the states from this uniform system. Further, the state legislators do not feel that a license with an electronically verifiable SSN (social security number) decreases the instances of a driver's SSN being illegally appropriated. In this age of technology, it will only be a matter of time before anyone could read a license electronically.

The overwhelming majority of states do not have centralized driver's license issuance. That is, driver's licenses are issued at counters at multiple remote locations within the state. When drivers apply for their license, they receive the license from the location where they apply, rather than a temporary paper copy. The official license is not sent through the mail from one central location within the state. This means that states would not only have to assure computer access to the federal government at each remote location, but these remote sites must be able to produce these federally approved licenses. Setting up these systems in these remote locations is a considerable additional expense.

In addition to an enormous fiscal impact, there is an essential time component involved in verifying the social security numbers. State legislators emphasize the importance of customer service at the DMV and are concerned with numerous constituent complaints about long lines and time requirements. This would only serve to grossly exacerbate consumer complaints and waiting time at the DMV.

Further, this aspect of the regulation intrudes on state authority, yet again. Rhode Island, like many states, explicitly does not require an individual to submit a social security number to the driver registry.

In the case of non-citizens, there are currently 26 different INS documents that prove lawful presence in the U.S. Under the new federal guidelines, states would be forced to recognize and understand all 26 forms of ID, even in states with a low immigrant population, in order to verify that these are "acceptable" forms of identification.

Also, to comply with the guidelines and avoid exposure to anti-discrimination litigation, (pursuant to Title VI of the Civil Rights Act of 1964) state and local governments would be required to provide citizenship and status verification for every single person applying for a license. This would force states to undertake costly staff training and public information campaigns to educate consumers on the need to provide additional documentation.

We have several concerns relating to the Preliminary Regulatory Evaluation (PRE) issued by the National Highway Traffic Safety Administration, Office of Regulatory Analysis and Evaluation in June of 1998. Our analysis reveals several violations of the Unfunded Mandates Reform Act.

The PRE names only the American Association of Motor Vehicle Administrators (AAMVA) as a consultative force in promulgating the proposed rule. Neither state fiscal officers, legislators nor governors or their associations were consulted in determining the financial and political ramifications of the changes proposed within the proposed rule, as required in Title II of UMRA.

The severity of this omission can be drawn from recent actions on the part of the administration in revoking Executive Order 13083 on Federalism, which similarly was criticized for failure to consult with state government officials and their organizations and was suspended precisely for that reason. We remain concerned that NHTSA would offer the PRE and the proposed rule without consultation with state officials, their staff and their national organizations. Heightened public awareness of consumer fraud related to "identity acquisition," privacy concerns and efficiency of consumer service at the DMV makes the proposed rule difficult to implement politically. Consultation with elected state officials and their organizations is vital to successful implementation of the proposed rule.

 Additionally, we have substantial concern as to the cost estimates offered in the PRE. We are troubled by the selection of the states used to determine the estimate and the methodology used to determine the costs. After contacting seven states, NCSL believes strongly that the costs associated with implementation of the rule as proposed far exceed those contained within the PRE, are over $100,000,000 and therefore violate the UMRA. Further, unlike the NHTSA we have not attempted to extrapolate the costs estimated by these seven states to the nation as a whole.

In Section V of the PRE, NHTSA discusses the estimated costs to states for implementing Section 656(b) of the Immigration Reform Act. Its analysis is based on estimated costs submitted by the states of Delaware, Iowa, Montana, Utah and Wisconsin. NHTSA admits that the AAMVA was the only state organization consulted in promulgating the proposed rule and no elected officials were consulted. Unfortunately, this means that only AAMVA members, i.e. state and provincial motor vehicle administrators in the surveyed states, provided the estimates of the proposed costs to implement the rule.

NCSL questions both the validity of the cost estimates provided by the five states surveyed, and the validity of using these five states as a basis for inferring the cost of the proposed rule to all 50 states, the District of Columbia, Puerto Rico, Guam and the Virgin Islands. The cost estimates submitted to us from the seven states we surveyed far exceed the numbers in the DOT cost evaluation.

A summary review of recent U.S. population changes from 1990 to 1997, and for the one year period from July 1996 to July 1997, reveal that four of the five states included in the AAMVA survey have not experienced large increases in overall population growth as compared to other states. The fifth state, Utah experienced a population growth due to a higher than average birthrate. None of the five states experienced large increases in population due to domestic migration or international migration.

The proposed purpose of the Immigration Reform Act was to "improve deterrence of illegal immigration into the United States" (1). Would it then not be more appropriate to select states for cost estimates that are assumed to be the most affected by immigration, both legal or otherwise? None of the states included in the cost estimate appears to be experiencing a large increase in migration into the state. The states included in the sample rank 26^th, 31^st, 33^rd, 41^st, and 51^st for the period 1996 to 1997 (see Attachment 1).

The following table illustrates the differences between the five states experiencing the largest increase in "net international migration" for the period July 1996 to July 1997, the latest period for which Census estimates are available, as well as the five states included in the PRE. While an estimate of the five states most affected by migration would also not provide a valid sample for estimating the national costs, inclusion of these states in an estimate would provide a more accurate reflection of the costs.

The following include some of the concerns we have as to the methodology used to determine national estimates for the cost of the proposed rule:

The national estimates, as developed in the PRE, cannot be substantiated and do not sufficiently address the costs associated with implementing the proposed rule. The PRE estimates a rosy, best case scenario, and substantially underestimates the cost of the proposed rule to states.

On its face, the estimated national cost is grossly underestimated. NCSL, in consulting seven states on the cost of the proposed rule, estimates that the cost of implementing the proposed rule far exceeds the one-time and annual costs listed in Table 3 of the PRE.

 According to the Unfunded Mandate Reform Act of 1994, if a mandate imposed on states exceeds one hundred million dollars in cost, it is in violation of the act. The lowest estimate of initial cost to a state that NCSL received was $3.1 million for Washington state. This is only initial cost. Even if every state could set-up a computer system for as little as $3.1 million, that still comes to more than $150 million for all fifty states--just to set up the computer system. This does not account for lost time, retraining employees and the cost associated with issuing new types of licenses. It also does not account for the cost of a public education campaign to inform residents of federally mandated forms of identification. These regulations amount to nothing more than an unfunded mandate. States could then bring suit against the Federal government for violation of UMRA. Needless to say, NCSL opposes them on that basis alone.

Title II of UMRA requires that agencies prepare a written statement containing a qualitative and quantitative assessment of the anticipated costs and benefits of the federal mandate. These statements are required to include:

• an analysis of the cost and benefits associated with the federal mandate,
• the extent to which such costs to State, local and tribal governments may be paid with federal financial assistance, and
• the extent to which there are available federal resources to carry out the intergovernmental mandate.

It is quite apparent that the NHTSA has violated UMRA in the spirit of the law in issuing cost estimates that are not methodologically sound nor an adequate reflection of the costs that will befall states in complying with the federal mandate.

Further, UMRA requires agencies "to the extent permitted in law, [to] develop an effective process to permit elected officers of State, local, and tribal governments (or their designated employees with authority to act on their behalf) to provide meaningful and timely input in the development of regulatory proposals containing significant Federal intergovernmental mandates." The NHTSA consulted only the AAMVA in promulgating the proposed rule. The AAMVA does not include elected officials. While state motor vehicle administrators may be best suited to administer the mandate, their interests in determining the cost of such programs must be tempered by input from state elected officials who appropriate their budgets.

Recommendations

 NCSL strongly objects to the rule as proposed.

In summary, state legislators are concerned and dismayed on many levels. The federal government obviously has not taken into consideration the overwhelming cost actually associated with these proposed regulations, and has therefore failed to notice the illegality in regards to the Unfunded Mandate Reform Act of 1994. The federal government has not considered the time and administrative impact that will negatively affect the issuance of driver's licenses to the public. These regulations intrude on and preempt state authority.

NCSL expects the NHTSA to remedy the concerns NCSL has with both the cost estimates and the process of consultation. NCSL firmly believes that the NHTSA must reevaluate the cost to states to implement the proposed rule.

NHTSA should:

1. withdraw the regulation until there can be a meaningful consultation with state and local elected officials and their national organizations.
2. issue an estimate that includes a broader sample of states in its estimate,
3. include those states that have had significant increases in international and domestic migration,
4. include states with large population increases; and
5. if the cost of the proposed rule proves to be above the $100 million threshold determined by the Unfunded Mandate Reform Act, the Administration should immediately seek funding from Congress to implement the federal mandate or join us in seeking to repeal the proposed rule.

We strongly urge the NHTSA to move expeditiously to issue a revised cost estimate on the proposed rule. We remain confident that a re-estimation of the costs will find that the proposed rule is well above the threshold set by UMRA and would burden states with a sizable unfunded mandate.

We also urge the NHTSA to seek other voices; particularly those of elected state officials, their staff and their national organizations, in reevaluating the cost of the proposed rule. The Preliminary Regulatory Evaluation reflects only the views of the AAMVA and there should be by a more inclusive discussion of costs with state elected officials.

Given our great concerns over this NPRM, NCSL would like to request a meeting with you to discuss these regulations further. Please contact either Dawn Levy or Sheri Steisel at NCSL's Washington office at 202-624-5400.

Thank you for your attention.

Sincerely,

Carl D. Tubbesing
Deputy Executive Director
The National Conference of State Legislatures