OPINION

This consolidated appeal concerns the constitutional validity of an initiative which was enacted by the voters as “The Alaska Homestead Act.”1

*2In two different court actions, the initiative was attacked as unconstitutional. After the superior court ruled in both cases that the initiative was invalid,2 the cases came before us on an expedited basis, before the general election at which the initiative could be considered. After hearing argument, we determined that the difficult constitutional questions presented in these cases could not be decided properly within the short time remaining before the election. We ordered that the initiative should be submitted to the electorate, reserving our determination on the merits for later decision. At the election, the initiative was approved, thus requiring that we decide the constitutional challenges.3

Four main issues have been presented to us:4

1. Does the initiative make an appropriation, which is prohibited by the state constitution?

2. Does the initiative amount to special legislation which is prohibited by the Alaska Constitution?

3. Does the initiative amount to a denial of equal protection of the laws under the state and federal constitutions?

4. Was ch. 181, SLA 1978 substantially the same measure as the initiative, thus voiding the initiative?

We have determined that the first issue is dispositive, and thus we do not consider the rest.

According to Alaska’s Constitution, the power of initiative “shall not be used to . make or repeal appropriations.” 5 The superior court found that the initiative for an act entitled “The Alaska Homestead Act,” popularly known as the Beirne Initiative,6 constituted an appropriation. For reasons that follow, we affirm that decision.

The Beirne Initiative makes available 30 million acres of state land to residents of Alaska.7 Under the terms of the proposed statute, certain Alaska residents are eligible to receive specified amounts of land, the exact amount of land being determined by the person’s length of residency: a resident of three years may receive up to 40 acres, a resident of five years up to 80 acres, and a resident of ten years up to 160 acres.8 Section 2 of the Initiative contains a part, Eligibility and Application,9 which indicates what a person must do to receive a grant:

A person . . . shall receive a homestead grant upon

(1) recording a written application in the form prescribed by AS 34.15.045 in the recording district in which the . land is located; and

(2) filing with the director the following materials:

(A) a copy of the recorded application,

(B) proof of residency,

(C) a $100 filing fee, unless the applicant submits a title search as provided in (b) of this section, in which case the filing fee is $75.

*3The applicant must also publish notice containing a description of the parcel and stating that application has been made under the Alaska Homestead Act to obtain title to the parcel.10 Finally, a survey of the land must be provided within five years after the grant is received or it is forfeited.11

The people, when approving Alaska’s Constitution, retained in article XI, section I, power to “propose and enact laws by the initiative, and approve or reject acts of the legislature by the referendum.”12 The right of initiative and referendum, sometimes referred to as direct legislation, should be liberally construed to permit exercise of that right.13 The reason given by the Supreme Court of Arizona for this rule of interpretation applies to Alaska:

It is, of course, a mere platitude to say that the people are the supreme power in our system of government. The history of our constitution and its adoption . shows beyond the possibility of contradiction that the people themselves deliberately and intentionally announced that, by its adoption, they meant to exercise their supreme sovereign power directly to a far greater extent than had been done in the past .

Whitman v. Moore, 59 Ariz. 211, 125 P.2d 445, 450-51 (1942), overruled on other grounds, Renck v. Superior Court, 66 Ariz. 320, 187 P.2d 656, 660-61 (1947).14

The people did not, however, choose to retain unlimited powers of initiative. Rath*4er, in section 7 of article XI, they declared that certain types of laws could not be enacted by initiative:

Restrictions. The initiative shall not be used to dedicate revenues, make or repeal appropriations, create courts, define the jurisdiction of courts or prescribe their rules, or enact local or special legislation.

The issue before us is whether the law proposed by the Beirne Initiative is, for purposes of section 7 of article XI, a law making an appropriation and, therefore, an illegitimate subject for initiative.

The basic principles for interpreting statutes apply to constitutions,15 and thus, to resolve this question, we shall turn to the language of section 7 construed in light of the purpose gf the provision.16 We are especially sensitive to the policy concerns embodied in constitutional provisions because a constitution is a document “unchangeable by ordinary means,” Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177, 2 L.Ed. 60, 73 (1803), which “must be considered as a living document adaptable to changing conditions and circumstances unanticipated at the time it was written.” Warwick v. State ex rel. Chance, 548 P.2d 384, 391 (Alaska 1976) (footnote omitted).17

Appellants argue that “appropriations” refers only to statutes setting aside a specific amount of money and designating it for a particular use. Appellees argue that “appropriations” includes statutes that set aside a specific amount of lands and direct it be given away in the manner required by the Beirne Initiative. The statute proposed by the Initiative gives away — to any resident of three or more years who will conduct a survey, file two papers, and pay a nominal filing fee — public assets in the form of state land. The proposed statute imposes no obligations on the applicant after he or she receives the land — the applicant need not homestead the land or live on it at all. The applicant may sell the land after one year and during that one-year period, the applicant incurs no tax liability.

The language of section 7 prohibits initiatives for the purpose of making appropriations. Though most state constitutions with referendum and initiative provisions have some limitation relating to appropriations, Alaska’s appropriation limitation is worded more generally than that of most other states.18 We found no decisions ex*5amining whether the term “appropriations” in the context of limiting the subject of initiatives prevents initiatives from appropriating land, or applies only to money.19 The term “appropriations” is sometimes used in the context of selecting land and designating its use,20 though appellants are correct that the usual context of appropriations is setting aside an amount of money and designating it for a particular use.21

*6Appellants, by defining appropriations as exclusively referring to money, implicitly agree that if the Beirne Initiative granted each Alaskan resident, depending on the length of residence, a sum of money, it would be an appropriations initiative and hence illegitimate. Does the fact that the proposed statute gives away land, rather than money, permit it to be enacted by initiative?

Appellants emphasis a change that occurred in the writing of the section restricting the subject of the initiative and referendum.22 The version proposed by the Committee on Direct Legislation stated:

Section 5. Neither the initiative nor referendum may be used as a means of making or defeating appropriations of public funds or earmarking of revenues nor for local or special legislation. Emergency acts are not subject to referendum.23

The Committee on Style and Drafting eliminated the phrase “of public funds” because:

We felt “of public funds” was not necessary because the only appropriations with which the state could deal are public funds anyway . . ,”24

The words “of public funds,” like the word “appropriations,” may convey different meanings. In one sense, funds may refer to money, while, in another sense, it refers to assets generally.25 Even if the *7initiative provision referred to appropriations “of public funds,” the issue would still be whether public funds refers generically to the state’s assets or only those assets in the form of money.26 We have concluded that by the term “appropriations,” article XI, section 7 prohibits an initiative whose primary object is to require the outflow of state assets in the form of land as well as money.

Victor Fischer writes of the complex of concerns that produced the various limitations on the subject matter of an initiative:

Further restrictions on direct legislation were added after expressions of concern about its potential use affecting fiscal legislation, local laws, the judicial system, and other “critical” areas. The convention reduced from three to two years the time limits during which an initiative could not be repealed by the legislature, and it provided that the initiative could be amended by the legislature at any time. These moves were considered necessary to assure consistency of initiative enactments with other laws of the state and with the general public interest as seen by the state legislature. The limitations upon direct popular legislative authority were a compromise designed to reserve basic authorities to the people while protecting the state against rash, discriminatory, and irresponsible acts.

V. Fischer, Alaska s Constitutional Convention 80-81 (1975) (emphasis added).

Initiatives for the purpose of requiring appropriations were thought to pose a special danger of “rash, discriminatory, and irresponsible acts.” The delegates were influenced by the experience of other states whose constitutions placed no restrictions on the subject matter of initiatives. They adopted the appropriations restriction to avoid the bad experiences of those states.27

The delegates wanted to prohibit the initiative process from being used to enact give-away programs, which have an inherent popular appeal, that would endanger the state treasury. A rather lengthy statement by Delegate Taylor 28 explains the delegates’ concerns:

Now in practically all the states that have initiative and referendum there are certain limitations put upon the matters that can be acted upon by those measures. Now appropriations are not subject to the initiative or the referendum. Some states made a great mistake by not restricting the initiative measures and allowed pressure groups to gather great numbers of signatures to a petition and that petition would require the expenditure of large amounts of money, perhaps a great deal more than the state could possibly afford and sometimes they would also initiate some legislation to raise money, a revenue measure and then directed *8that the proceeds of that measure would be utilized for a particular purpose. In other words, it took the making of revenue measures and expenditure of the funds away from the legislature and in some instances the governmental functions and governmental institutions suffered a great deal. And it was necessary within as short a time as possible to undo the damage that has been done.

Of course, if the proper safeguards are not put around the type of legislation that can be initiated by the people, as I said before, they can do a lot of harm. There was one in California that within a year they found out it was bankrupting the state, and they had to get out another initiative and do away with the first one. Colorado had the same experience, and the State of Washington, because they were levying taxes under those bills and directing where these taxes were going, and the State of Washington in a period of about eighteen months found themselves with not only losing a 60,000,000 dollar surplus that it had in the treasury but also 120,000,000 dollars in the hole. Colorado was about the same way.29

In Alaska, land is a primary asset of the state treasury. No other state government owns as much land.30 We cannot imagine the delegates’ concern over initiatives which depleted, for example, the Colorado treasury of its prime asset, public monies, coexisting with approval of an initiative which depletes Alaska’s treasury of its prime asset, public land.

The outflow of dollars from the state treasury is significant, not in and of itself, but because it represents an expenditure of the state’s assets. We see no rational set of policy concerns that would prohibit an initiative from giving away $9,000,000,000 but would permit it to give away 30 million acres, valued at that sum.31

Appellants attempt to distinguish this grant of land from a grant of money in two ways. First, they argue that private land ownership is a desirable policy goal from which the state will reap benefits, especially in the form of future tax revenues. Second, they argue that the cost of surveying the land may be substantial.

The benefits or lack thereof from a policy of private land ownership misconceives the issue. The question is whether this policy may be implemented through initiative, rather than through legislative action.32 The restrictions on permissible subjects for direct legislation represent “a recognition . that certain particularly sensitive or sophisticated areas of legislation should not be exposed to emotional electoral dialogue and impulsive enactment by the general public.” Stewart, The Law of Initiative Referendum in Massachusetts, 12 N.Engd.L.Rev. 455, 461 (1977) (footnote omitted). The danger with direct legislation relating to appropriations is that it “tempt[s] the voter to [prefer] his immediate financial welfare at the expense of vital government activities.” Note, Referendum: The Appropriations Exception in Nebraska, 54 Neb.L.Rev. 393, 394 (1975). Cf. Brown v. Ward, 593 P.2d 247 (Alaska 1979). The lure of an immediate grant of land poses the same temptation as an immediate grant of money. Both decisions are the kind that require the reasoned deliberation characteristic of legislative actions.

*9The fact that a survey may be costly does not change the essential nature of the Alaska Homestead Act as an appropriations initiative. The applicant pays the surveyor; 33 no compensation or service is rendered to the state.34 The stated purpose and effect of the Initiative on the state treasury is still an expenditure of state assets in the form of public lands.

Thus, the Alaska Homestead Act would substantially deplete the state government of valuable assets just as surely as an initiative allotting to residents of specified years large sums of money. In the same manner, it constitutes an appropriation and hence may not be enacted by initiative.

AFFIRMED.

concurring.

Though I am in accord with the majority’s views regarding the nature of the Alaska Homestead Act1 as an appropriation which may not be enacted by the voters through the initiative process, I wish to add my separate views concerning appellees’ equal protection arguments.

The Alaska Homestead Act, popularly known as the Beirne Initiative, provides grants of forty acres of state land 2 to Alaska residents3 who have resided continuously in the state for three years preceding application for land. Residents of five years duration are eligible to receive two forty-acre grants, and a person who has resided continuously within the state for ten years is entitled to four forty-acre parcels. Regardless of the length of time of residency, the Homestead Act provides that an individual is entitled to not more than one grant of forty acres per year, and no person may accumulate more than 160 acres combined total under the act.

I have concluded that these durational residency provisions violate the equal protection clause of the Alaska Constitution4 by infringing on the right to travel to this state and make one’s home here. A review of the case law illustrates the important nature of the right to travel protected by the state constitution.

*10The protection extended by the Alaska Constitution to new residents of Alaska differs markedly from the level of protection afforded by the United States Constitution to new residents of a state against discriminatory treatment based solely on the duration of residency.5 In the past, this court consistently has subjected durational residency restrictions to strict scrutiny under the equal protection clause of the Alaska Constitution. This intensified review was premised on our conclusion that “the right to interstate travel is itself a fundamental right and any classification which serves to penalize the exercise of that right must be subjected to strict judicial scrutiny.” 6

State v. Van Dort, 502 P.2d 453 (Alaska 1972), was the first case to challenge a durational residency requirement under Alaska law. In Van Dort we considered a seventy-five-day residency requirement for voting in state elections and found it to be unconstitutional. We stated there:

It is our reading of [Dunn v. Blumstein, 405 U.S. 330, 92 S.Ct. 995, 31 L.Ed.2d 274 (1972),] that all durational residency requirements are prima facie invalid as in contravention of the equal protection clause because they penalize the right to travel and the right to vote in elections on an equal basis with other citizens in the jurisdiction.

Id. at 454. The state’s argument that a seventy-five-day residency requirement for voting was justified by the compelling interest in guaranteeing that only bona fide residents participated in the election was rejected. Even recognizing Alaska’s uniqueness in size, population and geography, its special economic, cultural and social problems, and the communication deficiencies between different parts of the state, we found that a less restrictive alternative was available to accomplish the state’s objectives in administering proper elections. Thus, a thirty-day residency requirement expressly was approved in Van Dort; however, the seventy-five-day period challenged in the case was struck down since that particular classification served no compelling state interest.

In State v. Wylie, 516 P.2d 142 (Alaska 1973), we struck down certain state personnel regulations which gave an absolute hiring preference for state employment to per*11sons who had resided in Alaska for at least one year.7 We held in Wylie that the right of interstate travel is itself a fundamental right under the state constitution and that any classification which serves to penalize the exercise of that right must be subjected to strict scrutiny. At the time, this holding was based in part on our reading of the United States Supreme Court’s opinion in Dunn v. Blumstein, 405 U.S. 330, 338, 92 S.Ct. 995, 1001, 31 L.Ed.2d 274, 282 (1972), in which the Court had expressed its similar view that the right to travel was a “fundamental personal right.” However, Wylie established that aside from its status under the federal Constitution, the Alaska Constitution unambiguously protects the right of interstate migration as a fundamental interest such that strict judicial scrutiny must be applied to any burden on the exercise of that right.

Wylie significantly expanded the United States Supreme Court’s prior applications of the “penalty” concept to measures interfering with the fundamental right of interstate migration by holding that any differential treatment based on length of residency would be viewed as a penalty in light of the fundamental nature of the right of migration guaranteed by the Alaska Constitution. In Wylie, we held that state personnel rules which granted hiring preferences to persons who have satisfied the durational residency requirement penalized interstate travel. In contrast, though the precise limits of penalty analysis under the United States Constitution had not been explored fully at the time this court decided Wylie, the United States Supreme Court subsequently declined to classify as a penalty on the right to travel any burden other than the denial of basic necessities of life. See Memorial Hospital v. Maricopa County, 415 U.S. 250, 94 S.Ct. 1076, 39 L.Ed.2d 306 (1974).8

In State v. Adams, 522 P.2d 1125 (Alaska 1974), we again differed from the United States Supreme Court’s analysis of penalties on the right of interstate migration in holding that Alaska’s one-year durational residency requirement for the initiation of divorce proceedings in state courts violated the equal protection clause of the Alaska Constitution. We reiterated in Adams:

All durational residency requirements inherently infringe upon the fundamental constitutional right of interstate travel. *12Hence, all such requirements are prima facie invalid and will be countenanced only when they serve a compelling state interest. There need be no actual deterrence to interstate migration to actuate strict scrutiny under the compelling state interest test. In our view, the nature of the benefit withheld by the state is relevant only to judging the relative importance of the competing state interest, not to determining the applicable standard of judicial review.

Applying strict scrutiny to the state’s asserted public goals justifying the classification based on duration of residency, we found that the relationship between the goals and the means selected for furthering the objectives (the durational residency requirement) was substantially lacking. We held that the state’s important interest in protecting the integrity of the basic family unit simply was not furthered by the differential treatment of new residents since it did nothing to preserve marriages of persons who had been in the state for more than one year. The state’s interest in assuring the validity of its divorce decrees against collateral attacks on due process grounds, while substantial, could be satisfied by insuring that the domicile of the divorce-complainant was in this state. This reasonable, less restrictive alternative to the absolute requirement of residence of one year in the challenged divorce statute rendered the more restrictive standard less than compelling. The durational residency requirement thus failed to withstand strict scrutiny and was struck down.9

Following Adams, we decided the only Alaska right to travel case which has upheld a durational residency requirement in the face of strict judicial scrutiny of equal protection claims. Gilbert v. State, 526 P.2d 1131 (Alaska 1974), involved a challenge to the state constitutional requirement, article II, § 2, of three-year residency in the state and one-year residency in the election district for eligibility to seek legislative office. The residency requirement was analyzed only under the equal protection clause of the federal Constitution since a provision of the Alaska Constitution was directly attacked by the appellant in the case. Nevertheless, we determined to apply the strict scrutiny tier of review based in part on our own prior cases relating to durational residency requirements. Two of the state’s asserted justifications for the restrictions on candidacy for public office were found to be compelling: the interest of the state in assuring that those who govern it are acquainted with the diverse conditions, problems, and needs of those who are to be governed, and the interest of the electors in exposure of the potential candidate to the people during a period of time sufficient for the voters to become familiar with the candidate’s knowledge, character, and reputation. In upholding the classification, we found that no viable less restrictive alternatives to the durational residency requirements imposed by the state constitution existed to fulfill the compelling state interests.

This court’s most recent statement of the law relating to durational residency requirements was in Hicklin v. Orbeck, 565 P.2d 159 (Alaska 1977), rev’d on other grounds, 437 U.S. 518, 98 S.Ct. 2482, 57 L.Ed.2d 397 (1978). Hicklin involved an *13equal protection challenge to the so-called “Alaska Hire” law which limited eligibility for petroleum and pipeline related jobs to residents of the state who had been physically present in Alaska, with certain exceptions, for one year.10 In holding that the proper standard for review of the durational residency preference in hiring was the strict scrutiny tier of equal protection analysis which applies to such residency requirements “because they penalize those who have exercised their fundamental right of interstate migration,” 11 we expressly considered and rejected the “basic necessities” reasoning of Memorial Hospital v. Maricopa County, 415 U.S. 250, 94 S.Ct. 1076, 39 L.Ed.2d 306 (1974). As already noted, Memorial Hospital had addressed a challenge under the federal Constitution and had limited application of strict scrutiny to dura-tional residency cases in which the residency requirement penalized the right to travel by depriving the recent migrant of a basic necessity of life.12 We noted that “[w]e have never used this ‘basic necessity’ reasoning,” 13 and proceeded to subject the Alaska Hire law to strict scrutiny. The interests which the state sought to further through the hiring preference for one-year residents of the state in Hicklin were the reduction of the unemployment rate for bona fide Alaska residents and the cultivation of Alaska’s human resources through extraction of the state’s natural resources. We invalidated the durational residency requirements because the correlation between the classification based on length of residence and the goals of the challenged statute was tenuous and, in any case, the dura-tional residency requirement was not the least drastic means available for accomplishing the public goals of reduced unemployment and stabilization of the economy.14

In my view, under the test applied in Hicklin v. Orbeck, 565 P.2d 159 (Alaska 1977), rev’d on other grounds, 437 U.S. 518, 98 S.Ct. 2482, 57 L.Ed.2d 397 (1978), the three-year durational residency requirement imposed by the Beirne Initiative to qualify for land grants would fail. The general standard for judicial review of equal protection challenges arising under the state constitution was modified subsequent to the Hicklin decision, however. In State v. Erickson, 574 P.2d 1 (Alaska 1978), we abandoned the traditional two-tier approach to equal protection analysis which *14had been followed in all our prior cases 15 and adopted a new single test for evaluating equal protection claims under the Alaska Constitution. The new standard of review was explained in Erickson as follows:

Such a test will be flexible and dependent upon the importance of the rights involved. Based on the nature of the right, a greater or lesser burden will be placed on the state to show that the classification has a fair and substantial relation to a legitimate governmental objective. Where fundamental rights or suspect categories are involved, the results of this test will be essentially the same as requiring a ‘compelling state interest’; but, by avoiding outright categorization of fundamental and non-fundamental rights, a more flexible, less result-oriented analysis may be made.

Id. at 12. Our task in applying the equal protection test adopted in Erickson has three steps: first, we must ascertain what the purposes of the challenged legislation are and whether they are within the legitimate police power of the state; second, we must examine the means used to accomplish the legislative objectives and establish whether the means substantially further the legislative goals; and third, we must balance the importance of the state’s interest in the means actually chosen to accomplish the legislative purpose against the nature of the constitutional right which has been infringed.16

The Erickson analysis begins by considering the purposes of the challenged act, viewing the Beirne Initiative as a whole as well as the circumstances which surrounded its enactment. Since the act was passed as an initiative by the people rather than by the legislature there is no relevant legislative history of its passage to resort to. However, the Beirne Initiative makes its objectives clear in the statement of purpose which constitutes section 1 of the Homestead Act.17 The purpose of the act, quite simply, is to accomplish the transfer of thirty percent or 30,000,000 acres of the state’s vacant, unappropriated, and unreserved general grant lands into private ownership. The initiative finds that “individual land ownership is integral to the material well-being of the people and encourages more citizen awareness and involvement in the *15affairs of the state.” It further finds that the state unduly has delayed making public land available for private settlement and development purposes and enacts the grant program to facilitate the distribution of state land into private ownership.

The mechanism chosen for the transfer of land is a land grant initiated by filing an application which must have been recorded previously in the recording district where the land is located, together with proof of residency as required by the act and a $100 filing fee. The applicant is required to publish notice of his application and to provide a land survey within five years after the right to possession attaches. The grantee is required only to hold the land grant for one year before he “may extract timber or materials on a commercial basis, sell, subdivide, or otherwise dispose of the land . . . .” There is no requirement that the grantee actually enter upon the land and make improvements.18

Only persons who meet the minimum residency requirements of the Beirne Initiative may apply for grants of state land. A three-year resident is eligible for one forty-acre grant, a five-year resident may apply for two such grants, and a ten-year resident is eligible for a maximum of four forty-acre grants of state land under the act.19 The justifications advanced by the state in support of this differential treatment based on duration of residency are basically twofold. First, it is asserted that the durational residency requirements insure that the recipients of benefits under the Beirne Initiative have earned the right to grants of state land based on their contribution to the overall state welfare by voting and paying taxes for the required periods of time. Second, it is claimed that the durational residency requirements insure that the recipients of land grants are genuinely attached to the state since they already have demonstrated their willingness to remain in Alaska for a substantial period of time.

Judicial scrutiny of the closeness of fit between the means and permissible legislative objectives, and scrutiny of the relative necessity for utilizing the particular means selected as opposed to other means less restrictive of constitutional rights, vary in intensity with the character of the classification in question under State v. Erickson, 574 P.2d 1 (Alaska 1978). At the upper end of this flexible equal protection formula, where fundamental rights or suspect categories are involved, Erickson indicates that the results of this new test will be essentially the same as under the traditional strict scrutiny tier of equal protection doctrine which requires a “compelling state interest” in the means chosen. At the lower end of the scale, the intensity of scrutiny applied will be governed by the new, more demanding “rational basis” test articulated in Isakson v. Rickey, 550 P.2d 359 (Alaska 1976). In Isakson we held:

Under the rational basis test, in order for a classification to survive judicial scrutiny, the classification ‘must be reasonable, not arbitrary, and must rest upon some ground of difference having a fair and substantial relation to the object of the legislation, so that all persons simi*16larly circumstanced shall be treated alike.’20

Isakson also concluded that “we will no longer hypothesize facts which would sustain otherwise questionable legislation as was the case under the traditional rational basis standard.”21

The present challenge to the constitutionality of the durational residency requirements in the Beirne Initiative is the first post-Erickson case to come before this court on right-to-travel grounds. Thus, an important inquiry in this case is the determination of what standard of review should be applied. Our cases on this subject uniformly have expressed our opinion that the right of interstate migration is a fundamental constitutional guarantee which may not be infringed without showing a compelling state interest in retaining any durational residency requirement and the absence of less restrictive alternative means of accomplishing the state’s legislative objectives. The uniquely important status of right-to-travel protection in the Alaska Constitution reflects, in part, an awareness of the distinctive character of this state in attracting many new residents to participate in Alaska’s growth and expansion.22 The risk that longer-term Alaska residents would seek to insulate themselves from sharing the public benefits of development of the state with the influx of relative newcomers already has been confronted directly in Hicklin v. Orbeck, 565 P.2d 159 (Alaska 1977), rev’d on other grounds, 437 U.S. 518, 98 S.Ct. 2482, 57 L.Ed.2d 397 (1978). Significantly, this court in Hicklin stood fast in expressly repudiating the United States Supreme Court’s “basic necessities” limitation on applying strict scrutiny to right-to-travel cases and clearly stated that “it can no longer be disputed that [the right to travel into a different state and make one’s home there] is a fundamental right calling for strict scrutiny.” 23 In light of the consistent pronouncements by this court concerning the fundamental nature of the right to travel under our state constitution in prior cases, I believe there is a strong case presented in favor of continuing to apply the most stringent standard of review to durational residency classifications under the new, flexible equal protection test adopted in State v. Erickson, 574 P.2d 1 (Alaska 1978). However, even starting from the premise that the least intensive scrutiny available under Erickson applies, I do not think that the classification of Alaska residents into four groups of dissimilarly treated individuals based on duration of residency24 for purposes of disbursing *17grants of state land is “reasonable, not arbitrary” and “ ‘rest[s] upon some ground of difference having a fair and substantial relation to the object of the legislation, so that all persons similarly circumstanced [are] treated alike.’ ”25

The first rationale advanced for upholding the durational residency requirements of the Beirne Initiative is that they constitute a reasonable device for excluding from the benefits of the land grants those persons who have not made a substantial contribution to the state’s economy and general welfare. The state cites Starns v. Malkerson, 326 F.Supp. 234 (D.Minn. 1970), aff’d, 401 U.S. 985, 91 S.Ct. 1231, 28 L.Ed.2d 527 (1971), in support of its argument on this point. Starns upheld the University of Minnesota’s regulation creating an irrebut-table presumption that any person who had resided in the state for less than one year was a non-resident and could not qualify for a reduced resident tuition rate at the state university. The rationale advanced in Starns was that the one-year waiting period for reduced tuition was an attempt to achieve partial cost equalization between those who had and those who had not contributed to the state’s economy through employment, tax payments and expenditures within the state. The federal district court explained its holding as follows:

We believe that the State of Minnesota has the right to say that those new residents of the State shall make some contribution, tangible or intangible, towards the State’s welfare for a period of twelve months before becoming entitled to enjoy the same privileges as long-term residents possess to attend the University at a reduced resident’s fee.

Id. at 241. Assuming that the Alaska Supreme Court would uphold a similar one-year residency requirement for reduced tuition in this state’s university system, a premise that is not altogether inevitable under the new, “more demanding” rational basis test adopted in Isakson v. Rickey, 550 P.2d 359, 362 (Alaska 1976), substantial differences remain between the holding in the Starns case and the durational residency requirements of the Beirne Initiative.

A requirement of contribution through domicile in the state which funds an institution of higher education primarily out of tax revenues bears a relatively direct relationship to the rationale of cost equalization. The current residents of the state presumably carry the primary burden for providing the educational service and it is rational (under the deferential lower tier of federal equal protection scrutiny, at least) to require the new resident to pay more for a limited period of time to attend the state university in order to approximate the greater overall cost borne by the longer term resident for access to the same educational opportunity. In contrast, the distribution of land mandated by the Beirne Initiative constitutes the disposal of a permanent state asset the acquisition of which was unrelated to contributions of taxes and earnings by even the longest-term residents of Alaska. Of course, aside from the specif - *18ic funding sources of the state benefit in question, it is arguable that the longer-term resident has contributed more to the general welfare of the state over the course of his domicile in the state. However, in my view the asserted correlation between the contribution of the land grantee to the state’s economy and the varying periods of residency required to become eligible for each incremental forty-acre grant of land under the Beirne Initiative is on its face too tenuous to satisfy the standard of review adopted in State v. Erickson, 574 P.2d 1 (Alaska 1978). The parties to this action have made available no additional data supporting this rationale. Therefore, this court is left in the position of hypothesizing facts to sustain the classification on cost-equalization grounds, an exercise we expressly declined to undertake in Isakson v. Rickey, 550 P.2d 359, 362 (Alaska 1976). Further, if eligibility for land grants validly may be conditioned on participation in the state’s economy for such extended periods of time as the initiative requires, the state apparently would not be prohibited from distributing all of its public benefits (other than those protected by federal equal protection guarantees) proportionately to the length of continued residence in the state. I think that, upon careful evaluation, it becomes clear that the cost-equalization rationale advanced in this case in support of the dura-tional residency requirements accomplishes nothing more than the sanctioning of a bonus or reward for having been a resident of Alaska longer than certain other classes of persons. It would do violence to the meaning of the requirement in Isakson v. Rickey, id. at 362 (Alaska 1976), that the classification “ ‘must be reasonable, not arbitrary, and must rest upon some ground of difference having a fair and substantial relation to the object of the legislation’ ” if this court were to uphold the durational residency requirements of the Beirne Initiative on a cost-equalization rationale.

The second justification for the Beirne Initiative’s durational residency requirements advanced by the state is that they insure that state land grants made available by the act actually end up in the hands of those who have demonstrated their genuine attachment to the state and their willingness to remain in Alaska and adapt to its uniquely rigorous climate and frontier characteristics. I do not think this rationale is supported by the express provisions of the Beirne Initiative. Despite its title,26 the act is not a homestead law in the sense that the grantee must undertake to make improvements on the grant land or even to live on the land. In fact, under the terms of the initiative, a land grant applicant need not ever even visit the land acquired and can sell the land or otherwise dispose of it after holding it for one year. This scheme of land disposal may adequately serve the initiative’s purpose of transferring state land to private ownership, but it certainly does not require that the land is ultimately developed by those with any degree of attachment to the State of Alaska or that people with experience in “Alaskan living” utilize the land. The three-year durational residency requirement adds nothing to the assurance of these objectives.

Further, residence of three years duration is not reasonably required to insure that only actual residents receive grant lands. Whether or not the state has a justifiable concern that state land be distributed to bona fide residents of Alaska to the exclusion of non-resident applicants,27 requiring a minimum of three years actual domicile to establish bona fide residence status in this state simply bears no substantial relationship to the asserted purpose of the requirement. The Beirne Initiative itself includes other less onerous indicia of *19resident status which easily may be used to distinguish the resident from the non-resident.28

Given that the three-year durational residency requirement is not substantially related to the purpose of insuring that only bona fide residents with significant ties to the state receive the grant land, the five-year and ten-year requirements are even less relevant to this objective of the act. As stated before, the only conceivable relationship between the duration of residency in the case and the disposal of the state grant land is the notion that long-term Alaska residents deserve a reward for their continued residence. In my opinion, this connection does not meet the “fair and substantial relation” equal protection test of Isakson.29

dissenting.

I.

On the question of whether the initiative makes an appropriation, I respectfully dissent.

In Thomas v. Rosen, 569 P.2d 793, 796 (Alaska 1977), we quoted favorably from State ex rel. Finnegan v. Dammann, 220 Wis. 143, 264 N.W. 622, 624 (1936), which in turn states:

“An appropriation is the setting aside from the public revenue of a certain sum of money for a specified object, in such manner as the executive officers of the government are authorized to use that money, and no more, for that object, and no other.”

Courts in other jurisdictions have also defined “appropriation” for state constitutional purposes as applying to the use of money.1 But even more persuasive is the record of the proceedings of the Alaska Constitutional Convention. As originally proposed, the provision restricting the use of the initiative and referendum prohibited their use in “making or defeating appropriations of public funds.” 2 The floor discussion shows that the main concern of the delegates was that the expenditure of public money should not be subject to direct legislation by the electorate.3

It appears that the committee on style and drafting eliminated the words “of public funds” from the constitutional provisions because the state can only appropriate public funds. The committee chairman, Dele*20gate Sundborg, stated that those words were eliminated because the committee regarded them as redundant and unnecessary.4

In light of this history, I can find no sound reason for stretching the constitutional language to cover the distribution of land, a subject plainly not considered within the meaning of “appropriations” by our constitutional framers.

I would hold that the superior court erred in concluding that the initiative at bar amounted to a prohibited appropriation.

II.

My colleagues do not reach the question of whether the initiative violates the constitutional guarantee of equal protection of laws or the prohibition against special legislation. Because I would hold that the initiative does not amount to an appropriation, I must also express my views on these subjects.5

The Alaska Constitution, Art. XI, Sec. 7, provides that, “The initiative shall not be used to enact local or special legislation.” The homestead initiative is attacked as violating this provision. It is argued that the durational residency requirements of the initiative should be subjected to strict scrutiny as penalizing the right to travel. Alternatively, it is urged that the durational residency requirements do not bear a fair and substantial relationship to the purposes of the act.

I think the first argument is misplaced. It is true that laws which impinge upon the fundamental right of interstate migration, by denying the necessities of life or by penalizing the right to migrate, are invalid unless a compelling state interest justifies those laws.6

However, the initiative does not pose any obstacle to interstate migration. It does not deny to anyone the necessities of life or penalize a person for having migrated to Alaska. There is no fundamental right to receive land from the state to which one migrates. Certainly there are no cases which even intimate that such a right exists. The granting of state lands is a benefit conferred upon recipients of land grants, but it is not a penalty inflicted upon those to whom the grants are not made.

In this sphere of activity the state enjoys a large measure of autonomy. As the United States Supreme Court has observed:

“The governments of the States are sovereign within their territories save only as they are subject to the prohibitions of the Constitution as their action in some measure conflicts with powers delegated to the National Government, or with Congressional legislation enacted in the exercise of those powers.” Parker v. Brown, 317 U.S. 341, 359-60, 63 S.Ct. 307, 317-318, 87 L.Ed. 315 (1943).

The Alaska Constitution expressly permits the state to dispose of its lands. Art. VIII, Sec. 9, provides:

“Subject to the provisions of this section, the legislature may provide for the sale or grant of state lands, or interests therein, and establish sales procedures. All sales or grants shall contain such reservations to the State of all resources as may be required by Congress or the State and shall provide for access to these resources. Reservation of access shall not necessarily impair the owners’ use, prevent the control of trespass, or preclude compensation for damages.”

*21Thus the object of the initiative, a grant of land, is quite plainly within the lawful powers of the state.7

As noted above, this legislation does not require strict scrutiny. Therefore, the compelling state interest test does not apply. We have stated in our previous cases that where fundamental rights or suspect classifications are not in issue, we will employ a more deferential test.

“Examining both the legislative goals and the means used to advance them, we must determine whether the legislation bears a ‘fair and substantial relationship’ to legitimate purposes.” State v. Lewis, 559 P.2d 630, 643 (Alaska 1977).8

We have also noted that the test to be employed in determining whether a statute amounts to special legislation, prohibited by the constitution, is substantially the same as that applicable to non-suspect classifications which are challenged as violating the guarantee of the equal protéction of the laws. State v. Lewis, supra at 643.

I know of no principle of law which prohibits a state from distributing land to its residents upon the basis of how long they have resided in the state. The initiative before us parcels out state lands as an incentive to settle and develop the state, as well as to encourage citizen awareness of and involvement in the affairs of the state. Those who have spent more time in the state and who have made, it is presumed, a greater contribution to the state’s economy and well-being are given greater amounts of land. To insure permanent settlement of the land and to enhance economic growth, I think it is permissible for the state to base its land grants upon length of residence. This is a practical method of assuring that the land will go to those who have a more permanent attachment to Alaska than casual visitors or newcomers. In my view, the initiative does not violate the prohibition against special legislation nor does it deny to anyone the equal protection of the laws.

Because our test requiring a fair and substantial relationship between the legislative ends and the means used to achieve them is more exacting than the federal rational basis test, it follows that the federal requirements of equal protection of the laws are also satisfied.

It is my opinion that the initiative is constitutionally valid.