I dissent.
Public utilities have required some form of deposits since 1915 under purported standards comparable to those still used today and they understandably find it convenient to maintain the status quo. Thus the companies, and the Public Utilities Commission which approves their rules, appear genuinely perplexed when 56 years later a class of frustrated potential ratepayers question the validity of the current deposit requirement. The failure of the utilities and the commission is to take cognizance of two significant developments in the law during the past half century: the numerous constitutional decisions to prevent invidious discrimination in our society and the emergence of an innovative poverty law designed to assure the viability of our economic system.1
Petitioners desire to be consumers of telephone and gas and electric service and thus seek review of decision No. 76065 of the Public Utilities Commission which approved the “establishment of credit” rules of the Pacific Telephone and Telegraph Company (PT&T) and the Pacific Gas and Electric Company (PG&E). Petitioners contend that three of the rules create exceptions to the requirement of a cash deposit or other proof of *301credit which violate the equal protection clause of the Fourteenth Amendment. I agree, and am convinced that the Public Utilities Commission decision should be annulled.
The credit rules are set forth in the majority opinion and need not be repeated. Petitioners take issue specifically with the three credit rules which except from the requirement of a deposit or other proof of credit (1) persons owning real-property, (2) persons continuously employed by the same employer for two years, and (3) other persons able to establish credit “to the satisfaction” of the company, which category in practice includes private pensioners, employees of large corporations, and professionals. Petitioners contend that permitting such persons to escape the deposit requirement while compelling others to pay or meet alternative credit conditions (i.e., to obtain a guarantor or show prior uninterrupted service) is a denial of equal protection.
The majority of this court abdicate their responsibility to analyze the suspect categories by declining “to consider each subcategory separately” {ante, p. 296) and by merely painting with a broad brush characterized as “overall operation” of the rules. {Ante, p. 296.) The commission was obligated to consider each category separately, and there is no reason why this court should not do so on review. I can sympathize with the reluctance of the majority to face the issues raised by the petition, for the task of rationalizing the purported distinctions is obviously insurmountable. But sweeping the problem under the rug is unsatisfactory to the petitioners and unrewarding to the utilities which need appropriate guidance. Since the majority fail to meet the issues squarely, I shall undertake to do so.
In its least stringent application, equal protection of the laws requires a reasonable relation between classifications drawn under color of state law and the purpose for which the classifications are made. (See, e.g., Rinaldi v. Yeager (1966) 384 U.S. 305, 309 [16 L.Ed.2d 577, 580, 86 S.Ct. 1497].) The conceded purpose of the classifications within the telephone and gas company credit rules approved by the Public Utilities Commission is to reduce the bad debt losses of the companies by requiring a deposit from persons likely to be bad credit risks. Therefore, applying the traditional test of equal protection to the first two disputed credit classifications, the issue is whether there is a sufficiently rational correlation between land ownership or job stability and financial reliability, and conversely between lack of both land ownership and job stability and financial unreliability, so as to justify the imposition of more onerous economic requirements on persons without land and persons who change jobs.
It is clear that the two classifications are not necessarily related to credit reliability. A person of low income, who lives in rented lodgings and has *302not been at his job for two years, is often an excellent credit risk but he nonetheless may be required to make a cash deposit or meet other stringent credit requirements. If his income is such that he can afford modest monthly utility bills but not one or two $20 or $25 lump-sum deposits, he might be denied access to telephone or gas and electric service, or both. Therefore, the classification of nonlandowners and of persons without two years’ continuous employment is overinclusive with respect to its purpose of weeding out dangerous credit risks.2
But the classification is also underinclusive with respect to its purpose because many owners of land and many persons with job stability might nevertheless be poor credit risks.3 While financial resources are the primary requisite for land ownership, some persons with such resources are profligate and not necessarily responsible in meeting their financial obligations. As to mobility, a tenant with a term lease might he a more stable prospect than a property owner, especially if the latter’s equity is minimal. Job stability, though perhaps more closely related to credit reliability, is still an imprecise measure. Many factors unrelated to credit responsibility—such as ambition and advancement—might influence one man to change jobs and another to stay at the same position for two or more years. In short, the utilities appear to confuse ability to pay with credit reliability. Both land ownership and employment may be tangentially related to one’s ability to pay, but do not necessarily indicate that a person will be willing to pay his bills and hence be a good credit risk. If ability to pay is the relevant criterion, the utilities should have credit rules based entirely on gross income.
Because land ownership and job permanency are pragmatically related to income and race, the effect of the application of these empirically unsupported presumptions of credit reliability and unreliability is to impose the heaviest burden of the deposit requirement on the disadvantaged and the poor, while more affluent persons who may be credit risks are excepted. The anomalous result is that the poor who are users of utility services may *303be subsidizing others by bearing more than their fair share of the cost of bad debt losses through their involuntary payment of short-term, low interest capital in the form of cash deposits to the utility companies.4
As stated by a recent commentator, referring to credit, rules similar to those at issue in the case at bar: “Whether reliance on such crudely-defined categories reflects a belief that some groups never default, a fear that they would not tolerate the additional charge, or merely administrative convenience, these criteria hardly isolate with any precision those customers likely to default in payment of their utility bills. . . . [B]ecause of the deposit practices of many utilities, bad risks in the higher income groups pay no deposits while perfectly reliable users in lower-income must either deposit cash, undertake the often difficult task of proving their creditworthiness, or do without the utility’s service.” (Note, Public Utilities and the Poor: The Requirement of Cash Deposits from Domestic Consumers (1969) 78 Yale LJ. 448, 457-458.)
The commission, in sustaining the validity of the utilities’ establishment of credit rules, relied on evidence showing that utilities’ bad debt losses had been substantially reduced after they had instituted the rules. However, the fact that a deposit requirement imposed on persons not owning property and lacking job stability reduces bad debt losses does not demonstrate that land ownership and job stability are rationally related to credit reliability. As the utilities’ representatives admitted, a deposit requirement imposed on any group at random would reduce bad debt losses, perhaps to the same extent that the losses were reduced by the requirement imposed on persons without land and job stability. Without evidence to the contrary, it must be conceded that a deposit requirement imposed, for example, on the owners of real property might also reduce the bad debt losses incurred by the utility companies.
The commission’s decision refers us to a 1966 survey conducted by PT&T. The study involved 16,711 final accounts which had been written off as uncollectible during a 30-day period. Of this number, information as to customer background was unavailable in 5,556 cases. In the remaining group of uncollectible accounts, the statistics showed that 93 percent rented their living quarters and 71 percent had less than two-year job stability. From these figures, the commission deduced that persons who did not own *304land or remain at their jobs for two years were poor credit risks while persons with land or job stability were reliable. Hence, the commission concluded that the utilities’ credit rules were reasonably related to the reduction of bad debt losses.
The problem with the commission’s analysis is that the statistics relied upon are meaningless without related figures indicating the percentage of PT&T’s total number of ratepayers who rented their dwellings and had not remained at their latest jobs for two years. For example, if 71 percent of the total number of telephone company ratepayers held their jobs for less than two years, the fact that 71 percent of the uncollectibles also reflected that characteristic would lack significance.
The third contested credit classification excepts persons who establish credit “to the satisfaction” of the utility from the requirement of paying a deposit or otherwise establishing credit. If the first two classifications devised by the utilities to distinguish between good and bad credit risks are crudely defined, the third is not only impermissibly vague but its admitted application is wholly incomprehensible. As applied by the utilities, this third classification distinguishes between private and social security pensioners, employees of large and small corporations, and professionals and nonprofessionals. Neither the telephone company, the gas and electric company, or the commission offer any explanation to justify the distinctions drawn between private and social security pensioners and between employees of large and small corporations. No evidence is offered to show that the decision to favor one group over the other was even remotely related to the avowed purpose of the credit rules: to isolate bad credit risks and require them to pay a deposit or meet other requirements. As applied by the utilities, the rules might require a social security pensioner with an excellent credit background to pay a deposit, while a private pensioner with a history of irresponsibility would be excepted. The commission has suggested no defensible justification for distinguishing between utility customers on the basis of the size of their corporate employers or the source of their pension checks, when the stated purpose is reduction of bad debt losses. The application of this credit rule is wholly devoid of reason, and any effort to sustain it would be stultifying. The majority of this court, understandably, make no effort to conjure up a rationale for this classification.
The commission suggests some basis for differentiating between professionals and nonprofessionals. Professionals, who must meet extensive training and licensing requirements in order to practice within the state, are likely to remain here, the commission explains. Although it may be true that professionals as a group are less likely to leave the state, the classification is, at best, only marginally related to credit reliability. A party likely to remain within the state will be available to legal process, but he is not *305necessarily a good credit risk. Furthermore, the fact that many professionals are self-employed would seem to be an offsetting factor, since the income of a salaried employee is less likely to fluctuate and is more accessible to creditors.
Applying the traditional test of equal protection, I believe that the first two classifications based on land ownership and job stability are of doubtful constitutionality, while the application of the third is a clear violation of equal protection. But petitioners argue persuasively that a more stringent test of equal protection than the rationality requirement should be applied to the land ownership and job stability classifications.
It has become commonplace constitutional doctrine that “suspect classifications” and classifications affecting fundamental interests are subject to “special scrutiny” by the courts, and these classifications are sustained only if they are shown to be necessary to promote compelling governmental interests. (See, e.g., Shapiro v. Thompson (1969) 394 U.S. 618, 634, 638, 660 [22 L.Ed.2d 600, 615, 617, 630, 89 S.Ct. 1322]; Williams v. Rhodes (1968) 393 U.S. 23, 30-31 [21 L.Ed.2d 24, 31-32, 89 S.Ct. 5]; Harper v. Virginia Bd. of Elections (1966) 383 U.S. 663 [16 L.Ed.2d 169, 86 S.Ct. 1079]; Skinner v. Oklahoma (1942) 316 U.S. 535, 541-542 [86 L.Ed. 1655, 1660, 62 S.Ct. 1110]; Castro v. State of California (1970) 2 Cal.3d 223, 235-244 [85 Cal.Rptr. 20, 466 P.2d 244]; Note, Developments in the Law—Equal Protection (1969) 82 Harv.L.Rev. 1065, 1120-1131.) Petitioners contend that the utilities’ classifications must meet a compelling interest test because they are based on the suspect criterion of wealth and because they impair the fundamental interest of persons to obtain service from government-regulated utilities without meeting arbitrary preconditions.
In Harper v. Virginia Bd. of Elections (1966) supra, 383 U.S. 663, 668 [16 L.Ed.2d 169, 173], the United States Supreme Court made it abundantly clear that “[l]ines drawn on the basis of wealth or property, like those of race [citation], are traditionally disfavored.” (Also see Lee v. Habib (1970) 424 F.2d 891, 898 [137 App.D.C. 403].) And in McDonald v. Board of Election (1969) 394 U.S. 802, 807 [22 L.Ed.2d 739, 744, 89 S.Ct. 1404], the court reiterated that “a careful examination on our part is especially warranted where lines are drawn on the basis of wealth or race [citation], two factors which would independently render a classification highly suspect and thereby demand a more exacting judicial scrutiny.” Petitioners contend that classifications based on land ownership and job stability are fundamentally grounded on differences in wealth. Indeed, both land and employment are, in essence, sources of wealth, and land ownership cannot be achieved without the acquisition and expenditure of *306financial resources. Thus, the disputed credit rules may be viewed as tests of wealth rather than of financial responsibility.
I must assume that the commission would have invalidated the exceptions to the deposit requirement for landowners and persons with job stability if it had judged them by the standard applicable to classifications based on wealth. Neither classification could be demonstrated to be necessary to promote a compelling interest. The classifications are both over and under-inclusive and their burden falls upon the poor generally rather than upon bad credit risks specifically. To justify the discriminatory effect of the classifications, the utilities would have had to show that the interest in reducing bad debt losses was compelling and that other means less violative of the equal protection clause were not available. Neither of these showings could have been made. Most businesses treat bad debt losses as one of the inevitable costs of doing business and spread the loss among their paying customers.5 The utilities could do likewise with only slight burden, if any, to their ratepayers. Furthermore, even if reduction of bad debt losses were deemed a compelling interest, the utilities can achieve the result through other means less likely to discriminate against the poor. For example, utilities might require bad credit risks who have a history of default to pay their utility bills more frequently than other ratepayers; or, if the companies preferred a deposit requirement, the requirement could be limited to those customers with a previous history of unreliability in paying utility bills.
Petitioners contend also that the utility companies’ classifications fall within the second prong of the compelling interest standard—the classifications affect the fundamental interest of persons to enjoy equal access to utility service. In Harper v. Virginia Bd. of Elections (1966) supra, 383 U.S. 663, the Supreme Court recognized a fundamental interest in the franchise even though there was no constitutional right to vote in state elections, and the court viewed the poll tax with “special scrutiny.” I do not suggest that utility services are equal to the voting franchise in priority, *307but such services are necessaries of life in this mechanized age.6 In Sokol v. Public Utilities Commission (1966) 65 Cal.2d 247, 255 [53 Cal.Rptr. 673, 418 P.2d 265], we recognized that telephone service is an “essential means of communication for which there is no effective substitute” and that access to telephone service affects freedom of speech. And in City & County of San Francisco v. Western Air Lines, Inc. (1962) 204 Cal.App. 2d 105, 129 [22 Cal.Rptr. 216], the court noted that “a public utility is the dedication of property to a public use. . . . ‘ “Public use . . . means the use by the public and by every individual member of it, as a legal right.” ’ ” Therefore, classifications affecting the access of the poor to utility service should be viewed with “special scrutiny,” given that the right to utility service is essential and that the denial of access inflicts hardship on those deprived. “The right to public utility services may not rank with the franchise, or with procedural due process in criminal cases, on a conventional scale of liberties, but to the poor nothing may be more immediately important than fair treatment by those who supply the needs of their daily existence.” (Note (1969) supra, 78 Yale L.J. 448, 463.)7
Classifications based on land ownership and job stability are not to be viewed with the same suspicion as those based, for example, on race. Nor do I believe that the interest in equal access to utility services is deserving of the same protection as the right to vote or the constitutionally protected right to procedural due process in criminal cases. Nevertheless, the classifications before us involve both the suspect criterion of wealth and the crucial interest of persons in equal access to utility service. This combination of factors should compel “special scrutiny.”
In a recent examination of the requirements of equal protection, the *308Supreme Court subjected state welfare residency requirements to “special scrutiny,” despite the absence of both a traditionally suspect basis of classification and a “fundamental” interest normally placed in the highest rank. In Shapiro v. Thompson (1969) supra, 394 U.S. 618, 634 [22 L.Ed.2d 600, 615], the court explicitly rejected a rationality test solely because the residency requirements imposed by the states affected the right of persons to travel from state to state. We would not broaden the application of “special scrutiny” in the instant case were we to apply it to utility company classifications founded on wealth and affecting the ability of persons to acquire modern day necessities of life.
For the foregoing reasons, I am convinced that the contested credit rules do not meet the standards imposed by the equal protection clause. The subclassifications erected in the application of the flexible category for credit established “to the satisfaction” of the utilities are not rationally related to the avowed purpose of the credit rules; consequently, these subclassifications constitute an invidious discrimination against social security pensioners, employees of small corporations, and nonprofessionals. The other two classifications, imposing more stringent credit requirements on persons not owning land and lacking two-year job stability, are of doubtful constitutionality under a rationality test and clearly unconstitutional when subjected to “special scrutiny.” Therefore, the commission’s decision sustaining the validity of the utilities’ credit rules should be annulled.
In disapproving certain credit rules on equal protection grounds, I would not impose an “all-or-nothing” approach on the utility companies. A deposit need not be exacted from every utility customer, or not at all, in order to pass constitutional analysis. The Public Utilities Commission, in protecting ■ the public interest, should require only that credit rules, if any, bear a significant relationship to the purposes sought to be achieved by the imposition of the rules. If the object is to isolate bad credit risks and to require only from them deposits or other proof of credit reliability, the classifications established to achieve the purpose must be defined with sufficient precision to avoid invidious discrimination in the imposition of credit burdens and benefits.
Peters, J., concurred.
Petitioners’ application for a rehearing was denied April 29, 1971. Tobriner, J., and Sullivan, J., did not participate therein. Schauer, J.,* and Sims, J.,† participated therein. Peters, J., and Mosk, J., were of the opinion that the petition should be granted.
The public policy of the United States was stated in the preamble to the Economic Opportunity Act of 1964 (42 U.S.C. § 2701): “Although the economic well-being and prosperity of the United States have progressed to a level surpassing any achieved in world history, and although these benefits are widely shared throughout the Nation, poverty continues to be the lot of a substantial number of our people. The United States can achieve its full economic and social potential as a nation only if every individual has the opportunity to contribute to the full extent of his capabilities and to participate in the workings of our society. It is, therefore, the policy of the United States to eliminate the paradox of poverty in the midst of plenty in this Nation by opening to everyone the opportunity for education and training, the opportunity to work, and the opportunity to live in decency and dignity.”
“Justice in a society such as ours, a society marked by wide differences in wealth and power, requires a legal system that compensates for these differences. The law is above all a means of creating and protecting rights. What is so necessary is an enlarged conception of the rights of the poor and a changing conception of the role of law in providing, protecting and implementing these rights. We must disabuse ourselves of the myth that poverty is somehow caused by the poor. We must recognize that the law often contributes to poverty. We must understand that whereas the law for most of us is a source of rights, for the poor the law appears always to be taking something away. That we have to change. And those of us who have been given the temporary custody of our laws by the people must ensure that those laws and our courts treat all equally—rich and poor alike.” (Address, Justice for the Poor, by John N. Turner, Minister of Justice and Attorney General of Canada, North American Judges Association Conference, San Francisco, Dec. 1, 1969.)
Although discussing merchandise, rather than public utility services, Professor Dodyk, et al., in their book. Law and Poverty (1969) at pages 837-838, conclude that “Many low-income tamiles are quite capable of making regular payments. . . . Moreover, perhaps general market retailers can take steps to make it easier for low-income families to apply for and receive credit. Some retailers have already found that they can do so economically. Various community business organizations might consider ways of more actively encouraging low-income families to seek credit. . . .
“Increased competition for the patronage of low-income consumers would go a long way toward resolving many of the problems confronting them in the low-income market. Public policy should consider the various ways by which new entrants could be encouraged into these markets . . . .”
With all its vast multi-state resources, the Penn Central Railroad, currently in receivership, is less likely to pay utility bills for its California offices promptly and in full than the neighborhood grocer.
Both PT&T and PG&E rules provide for the return of the deposits to the customers after one year of continuous service without disconnection. PT&T pays interest at the rate of one-half of 1 percent per month and PG&E pays 5 percent per year.
PT&T presented evidence that it collects approximately $3,000,000 in deposits from over 100,000 customers annually. Although this amount may constitute but a fraction of the total revenues of PT&T, $3,000,000 is nonetheless a sizeable sum acquired at minimal interest cost.
We live in a society in which consumer credit is a major economic factor. At the end of 1945 consumer credit amounted to $5.6 billion; by 1967 the total amount was estimated at $92.5 billion. The growth has been four and a half times greater than the growth rate of our economy as a whole. (Dodyk, et al., Law and Poverty (1969) p. 819.) It seems strange that most commercial enterprises extend longer credit than utilities and few, if any, require cash deposits. Yet utilities operate with government-regulated rates assuring them of a profit.
Most businesses find expanding consumer credit to be rewarding. “Bank of America has about $2.6 billion in instalment credit outstanding, or approximately 17 per cent of its loan portfolio. These loans have a high rate of repayment—about $400 million per month. They are good income producers and have had a remarkably low loss record—less than one quarter of one per cent since the bank entered the consumer credit field in 1929.” (BankAmerica Corporation Annual Report 1970, p. 12.)
Petitioner Wood, a 76-year-old pensioner, testified before the commission and described the hardships he suffered while without utility service: “[Tjhere was no heat so I went to bed early. I had my own bed clothing, and after having had quite a bit of experience with cold weather, I took a small blanket and put it over my head like a parka. ... I ate some cold lunches. And there are a few greasy-spoon restaurants around that neighborhood so that is how I lived.”
A social worker testified with regard to telephone service: “I had one woman who lay on the floor for two days with a broken leg because nobody came in to check to find out why she hadn’t been seen or heard from. If there had been a phone, they might have been able to.”
Seeking to articulate a general formula to distinguish “fundamental interests” from others for purposes of the equal protection clause, a commentator has recently suggested “that a common thread can be found in the severity of the detriment imposed on a complaining party in ‘fundamental interest’ cases.” (Note (1969) supra, 82 Harv.L.Rev. 1065, 1130.) Applying that test to a person unable to qualify under the utilities’ credit rules and unable to afford the lump-sum deposit, it may well be that deprivation of utility service is a detriment so severe that the right to equal access to utility service is a “fundamental” right. (Cf. Allen v. Railroad Commission (1918) 179 Cal. 68, 88 [175 P. 466, 8 A.L.R. 249].)
Retired Associate Justice of the Supreme Court sitting under assignment by the Chairman of the Judicial Council.
Assigned by the Chairman of the Judicial Council.