Saturday, December 28, 2013

Hobby Lobby Part III-A—Does federal law substantially pressure employers to offer health insurance coverage in violation of religious obligations, even though there is no “Employer Mandate”?

Marty Lederman

The plaintiffs in Hobby Lobby and Conestoga Wood argue that federal law compels them to act contrary to their religious obligations, by requiring them to offer (and pay for and administer) employee health insurance plans that include contraception coverage. As I explained in my most recent post, that turns out to be a simple misreading of the law:  Although employee plans must include contraception coverage, the Affordable Care Act does not require that employers offer such plans to their employees, nor even impose substantial pressure upon them to do so.

I received several thoughtful responses to that post, questioning whether that conclusion can be right.  My correspondents have not taken issue with my reading of the law.  Nor have they argued that the law imposes a duty on the plaintiffs to offer health insurance plans.  They have suggested, however, that the law must have a very different practical effect than I surmised because, if I were correct, then most employers would be dropping their health insurance plans—something that is not happening and that the government does not expect to happen anytime soon.

My correspondents are of course correct about their factual premise:  Some large employers will drop their employee heath insurance plans in response to the ACA, but most will not.  Why?  Is it because the ACA or other federal law places substantial pressure on them to retain their plans?  In this post, I’ll try to grapple with that question and then, toward the end, discuss how it might affect the specific RFRA question at issue in the Hobby Lobby and Conestoga Wood cases.

As I will explain, the mere fact that most large employers might on balance choose to continue to provide employee health insurance plans, even once their employees would otherwise be eligible to purchase insurance on an ACA “exchange,” does not itself establish that federal law itself imposes a sufficiently direct, let alone “substantial,” degree of pressure on employers to retain such plans, so as to establish the sort of “substantial burden” on religious exercise that triggers RFRA scrutiny.  And more importantly, plaintiffs in these cases have failed to allege facts that would demonstrate that federal law substantially pressures them to retain their plans, and thus (allegedly) to violate religious obligations.

1.  The Basic Argument:  There is no “Employer Mandate,” and the cost of the ACA “assessment” is far less than the cost of providing employee health insurance

To recap the basic argument from my recent post

The core of the plaintiffs’ RFRA claims in the Hobby Lobby and Conestoga Wood cases (and in virtually all the other challenges to the HHS “Preventive Services” Rule) is that the federal government has put them to what Hobby Lobby calls an “unconscionable choice” between violating the law and violating their religious obligations.  According to the plaintiffs, the HHS Rule, in conjunction with the ACA, requires them to offer certain contraception insurance coverage to their employees . . . which is conduct their religions proscribe.

As I explained in my post, however, federal law does not actually impose a duty on the plaintiffs to offer such contraception coverage.  To be sure, the law does require that contraception, and several other specified health care services, be included in any health insurance plan the employers offer to their employees.  But the law does not require them to offer health insurance to their employees at all.  Instead, it gives them an alternative option—they can lawfully decline to offer such insurance to their employees.  In that case their employees would no longer be left in the lurch, because the ACA itself would make them eligible for affordable health insurance without regard to health status (e.g., without any penalty for preexisting conditions) on a government-created “exchange”—insurance that will be subsidized for most employees by government-provided premium tax credits and cost-sharing reductions.  

It is true that if a large employer, such as Hobby Lobby or Conestoga Wood, drops its health insurance plan, a provision of the ACA, 26 U.S.C. § 4980H(a), would require such an employer to make an “assessable payment” to the IRS.  But contrary to popular belief (and the assumptions of several courts of appeals), that payment—less than $2000 annually for each full-time employee—is not a penalty for violation of a legal duty; it is, instead, a tax to help defray the social cost of subsidizing the insurance purchased on the exchanges.  And that tax, moreover, is far less than the average cost to the employer of the premiums and administrative expenses for its existing insurance plan.  Therefore, if we look only to the ACA itself, and the HHS Rule, not only is there no duty to offer employees participation in a contraceptive-included plan, but most or all large employers would save money by abandoning their plans . . . and could do so without denying their employees the opportunity to purchase affordable insurance without regard to health status (something they could not do pre-ACA).

2.  The challenge from my correspondents

My correspondents have raised an obvious question—namely, if I am right, then why aren’t most large employers dropping their employee health insurance plans?  Doesn’t the empirical evidence suggest that there’s something fishy in my account?  This is the way one astute observer put the point:

Your claim seems refuted by the empirical reality that most large employers are continuing to choose to provide insurance, rather than dropping their plans and gaining the supposed cost “savings.”  This reveals that they don’t actually think they’ll be better off dropping their plans (likely because of how their employees would react), and yet now the ACA is pressuring religious employers to do so and to pay a large tax on top if they want to avoid violating their conscience.

And you further seem to suggest that their employees, or at least most of the employees, wouldn’t demand a significant raise in salary to offset the loss of this valuable (tax-deductible) benefit, because they’ll supposedly receive roughly the same benefit from the government through direct subsidies for insurance on the exchanges.

But if all that’s really true, then why on earth is any large employer in this country continuing to provide health insurance post-ACA?  If the ACA’s combination of government-subsidized insurance for employees plus government-imposed taxes on employers really is better for both the employer and the employee, then shouldn’t every large employer in the country that’s economically rational immediately drop its insurance plan (and force the taxpayers to bear the cost of their employees’ insurance)? 

Simply put, don’t the actual revealed preferences of most large employers in this country refute your theoretical suggestion that employers view their ability to offer an insurance plan as a burden rather than a benefit? 

Great question—and obviously of potential relevance to the RFRA calculus.  So let’s unpack what’s going on.

3.  Effect of the ACA

The ACA does at least three things that are pertinent here:  (i) it increases the number of individuals, including employees of large employers, who will be eligible to receive health insurance through Medicaid; (ii) it offers affordable health insurance, without regard to preexisting conditions, to employees whose employers do not offer such insurance—and significantly subsidizes that insurance for lower-income individuals; and (iii) it imposes the § 4980H(a) assessment on large employers who do not offer health insurance, which helps to subsidize the first two changes.

My correspondents have not quibbled with my calculation that the § 4980H(a) assessment will be much smaller than the amount a typical large employer now pays for the administrative costs and the costs of premiums for its employee health insurance plan.  Nor have they taken issue with my claim that those employers’ employees would now have access to affordable health insurance, without regard to preexisting conditions, if the employer did not provide coverage, whereas they did not have that opportunity before the ACA. 

Therefore, if we look only to the ACA itself, and the HHS regulations under it, large employers would, in fact, save money by dropping their health-insurance plans.  (Notably, neither Hobby Lobby nor Conestoga Wood has alleged that any of the foregoing would not be true as applied to them.)

To be sure, not all employers will in fact drop their plans—nor is that something that Congress assumed or desired when it enacted the ACA.  Indeed, employers were not dropping their employee health insurance plans prior to the ACA, i.e., before the § 4980H(a) assessment kicks in (in 2015)—a phenomenon that was a function of pre-existing incentives and market conditions, which I’ll discuss below.  Among those who have studied the issue, however, there appears to be widespread agreement that the changes in the ACA itself will, on the whole, push employers in the direction of dropping their plans, even taking into account the § 4980H(a) assessment.

There is also a consensus among those who have studied the question that that “[t]here is clearly a tremendous amount of uncertainty about how employers and employees will respond,” such that there is no way of predicting with certainty how many employers will drop their plans because of the ACA—the models are simply too complex, and depend upon too many assumptions and predictive variables.  The Congressional Budget Office, for instance, estimates (see footnote e) that 11 million employees will lose their employer-provided coverage (and thus obtain their insurance on an exchange).  But the number could be fewer than that—or, on some projections, more.   
Observers agree, however, that whatever the ultimate number might be, there won’t be a wholesale employer abandonment of employee health insurance plans.  Why not?  Isn’t that, after all, what should happen if I were correct that the ACA assessment is almost certain to be far less costly than the premiums employers contribute to their employees’ health-insurance plans?

Well, no—it’s very unlikely that such a mass employer abandonment will occur.  But that’s not because of the ACA, or the § 4980H(a) assessment, in particular—instead, it’s the result of a complex array of factors that have deterred employers for decades from dropping their health-insurance coverage.  The ACA will alleviate, or counter, this deterrence to some degree, i.e., it will cause some employers to alter their historical practice, but not entirely.

4.  Pre-ACA Practice, and the Tax Exclusion for Employees

So let’s examine why many large employers have, for the past several decades, offered health-insurance plans to their employees.  It’s not because the law has required them to do so—it hasn’t; and it’s certainly not because of the § 4980H(a) assessment of the ACA, which won’t apply until 2015.

This widespread employer practice has been primarily attributable to two things.  The first is a federal government subsidy to employees:  Since at least 1954, employees have received a tax exclusion for the portion of their insurance premiums that are paid by their employers.  See 26 U.S.C. § 106(a) (“[G]ross income of an employee does not include employer-provided coverage under an accident or health plan.”).  Such employer-paid premiums are also excluded from calculation of employees’ social security and Medicare taxes.

In addition, at least in the pre-ACA world, the insurance the employees could obtain on the open market, using their own wages, was not nearly as attractive as employer-provider insurance:  Market insurance excluded (or charged substantially more to) many individuals with pre-existing conditions, and did not pool health risks nearly as efficiently as employer plans.  

OK, so that’s why employees have preferred to receive remuneration for their labor in the form of employer-subsidized health insurance rather than wages.  Even so, why have those employee preferences caused so many employers to offer health-insurance plans? 

Well, from the employer’s perspective, it has made little difference whether employees are paid for their labor in the form of wages or health-insurance premiums.  But their employees obviously prefer the latter option, for the reasons explained above (including that they are not taxed for the premiums paid by their employers).  Accordingly, making payments by way of subsidized employee health insurance, in lieu of wages, has long been a way for employers to satisfy employee preferences.  And in a market where that is the norm, an employer that failed to do so—one that continued to convey those dollars to employees in the form of wages, thus relegating its employees to purchase their own insurance on the private market—would be less attractive to many employees.  Because employers generally prefer to be competitive for the best employees, and because they are otherwise indifferent as to whether to remit wages rather than insurance premiums to their employees, most large employers have, for decades, gladly given employees what they prefer—lower wages than they otherwise would have received, in exchange for access to a valuable employer insurance plan.

5.  Impact of the ACA on established employer practices

The ACA makes employer-provided health insurance less attractive, or at least less preferable, for some employees.  The new law expands Medicaid eligibility, for one thing.  And it makes many other employees eligible for affordable insurance on an exchange—insurance that cannot discriminate on the basis of pre-existing conditions—if their employers decline to offer insurance.

These new options virtually eliminate one of the two principal reasons that employees have traditionally preferred compensation in the form of employer-provided insurance rather than wages—namely, the unavailability of affordable health insurance on the open market.  The other reason for that traditional preference, however—the tax exclusion employees enjoy with respect to employer-paid premiums—remains in place. 

Even so, purchasing insurance on the exchange under the ACA will in fact be a better option than employer-provided insurance for many lower-income employees, because the generous government-provided tax credits and cost-sharing to which they are entitled will exceed the value of the tax exclusion for employer premiums they now enjoy under their employer plans.  And thus it will be advantageous for employers whose work force consists primarily of such workers—and for the vast majority of those employees themselves— to drop their health insurance plan.  Accordingly, those are the employers that are most likely to drop their plans.

Some higher-income employees, on the other hand, will not be eligible for government subsidies on the exchanges.  Therefore, if their employers drop insurance coverage they will have to pay for health insurance using their own taxable wages, whereas previously their employers contributed tax-excluded dollars for a portion of the premium.  For these employees, the switch to exchange-provided-insurance might therefore be more expensive, and thus disfavored.

If a large employer wishes to remain competitive for such employees—or even simply wishes to ensure, for fairness reasons, that those employees do not suffer any loss in the real value of their compensation—the employer would have to do one of two things:  either (i) retain its health insurance plan, or else (ii) drop the plan but increase the wages paid to such employees to compensate them for the increase in their health-insurance costs.  An employer choosing the latter option will sometimes be able to increase such wages using money it has saved by dropping its insurance plan (even after accounting for the § 4980H(a) assessment).  In those cases, as in the cases of lower-income employees, dropping such insurance might be a win/win proposition for employers and employees alike.  But in other cases, the increase in wages that would be needed to fully compensate the higher-income employees will not be worth the candle to an employer, and such an employer may thus choose to retain its health-insurance plan, at least if it is in a competitive market for recruiting and retaining valuable employees.

This explains, in part, why many large employers may choose to retain their insurance plans, even after the exchanges become viable under the ACA. 

(For more details along these lines, see this recent article by researchers at the University of Michigan.)

But there are other obvious considerations that might affect employer choices, too.  Caution, for example.  Many employers may wait several months or years before deciding whether to abandon their plans.  Few employers will be eager to be the “first mover” within their marketplace in terms of dropping employee insurance.  They are more likely to opt to wait and see—to see how the exchanges operate in practice; what the effects of exchange-purchase would be on their employees; how employees react and what they demand (e.g., in collective bargaining) when an employer drops its plan; and, especially, how their competitors are reacting to the complex dynamics created by the new law.

Furthermore, in some corporations executives might be reluctant to abandon employee health insurance plans because the plans offer the executives themselves valuable benefits, beyond the legal minimum, that would be difficult or expensive to replicate on the exchange.  Executive self-interest, that is to say, might also be a contributing factor that pushes in favor of the status quo.

In sum, the general preference of most employers to retain their employee insurance plans is likely the result of a wide array of very context-dependent and fact-specific factors.  The most prominent factor is likely the long-established tax exclusion that benefits their employees.  The creation of the exchanges under the ACA, on the other hand, together with government subsidies and the Medicaid expansion, should push strongly in the direction of encouraging employers to drop their insurance plans.  And the § 4980H(a) assessment should push back--although not as far--in the opposite direction.  Employers will also be influenced by the elasticity of the relevant labor market; by their relative degree of caution and risk tolerance; by the nature and income levels of their workforce; by the viability and desirability of the exchanges over time; by the costs of their insurance plans; by their executives’ preferences for retaining such plains; by their competitors’ choices; and many more variables.

Therefore, even if it were true, as my correspondents have suggested, that most employers will retain their employee insurance plans (at least in the short run), that would not necessarily mean that most or all such employers are substantially pressured to do so, let alone that federal law creates such substantial pressure.  It might be, instead, that most employers have soft preferences to retain such plans, owing to a vast array of variables and considerations.

6.  How does this affect the RFRA analysis in Hobby Lobby and Conestoga Wood?

As I explained earlier, the basic argument that the plaintiffs have made under RFRA—that federal law will impose a duty upon them that is in direct conflict with their religious obligations—is based on a mistaken premise:  There is no such legal duty.

Recognition of this previously unappreciated fact should affect the way the Court views the “substantial burden” question under RFRA, and the RFRA calculus more broadly.  Even so, I explained in my last post, it would not necessarily defeat the RFRA claims altogether, because even without a legal mandate to offer contraception coverage, federal law might still place pressure upon the plaintiff employers to offer such coverage that is substantial enough to satisfy RFRA’s “substantial burden on religious exercise” prerequisite.  That is a teaching of cases such as Sherbert v. Verner and Thomas v. Review Board.

With respect ti this question, several things are worth noting in light of the discussion above about why many employers do not and will not drop their employee health-insurance coverage:

First, to the extent federal law is responsible for any pressure on employers to retain their plans, it is not so much by virtue of the ACA (which actually, on the whole, provides some incentives to employers to drop their insurance plans), but instead as a result of the 60+-year-old tax exclusion for employer premiums that the federal government affords to employees—that is to say, neither by operation of a penalty nor even a benefit offered to the employers, but instead by a benefit the government provides to other actors. 

Second, and relatedly, any pressure on employers to retain their insurance plans is not merely the effect of the employee tax exclusion law; rather, it is based on a complex array of factors, most of which are market-driven rather than compelled by law.  To summarize what I discussed above:

i.  Were it not for the pre-existing tax exclusion, most employees post-ACA would have little reason to prefer employer-provided health insurance to insurance they can receive on the exchange.  Indeed, for many employees, the latter will in fact be the preferred option, even taking the tax exclusion into account.

ii.  For some employees, however, exchange-purchased insurance will be more costly, because it will not be subsidized to the extent of the subsidy they have received by virtue of the tax exclusion for employer premiums under their current, employer-provided plans.

iii.  Employers that drop their health care plans can compensate some of employees in the latter category by raising their salaries.

iv.  Some employers who drop insurance coverage will still end up in the black, even with such salary adjustments, because of the significant savings they realize from not paying insurance premiums and administrative costs.

v.  Other employers, however, will have to pay so much in salary to keep those employees on an even keel that it would exceed what the employers save by dropping their plans.

vi.  That latter set of employers would then have at least two choices:  They could either keep their insurance plans, or they could drop their insurance plans and risk losing some of the employees in question to competitor employers who continue to provide employees access to tax-beneficial insurance plans.  Whether a particular employer is willing to accept that risk presumably would depend upon how competitive the relevant labor market is for such employees, and how important it is to that employer to retain such well-compensated employees.

That is to say:  The effect of federal law on employer decisions is quite indirect and attenuated, and dependent upon the preferences and predictions of nongovernmental actors—the employer’s employees and competitors, in particular.

Offhand, I can’t think of any RFRA cases, or cases from the relevant pre-RFRA Free Exercise jurisprudence, in which the alleged burden was the result of such an attenuated and multi-factored chain of causality. 

The closest analogy is probably the Court’s 1961 decision in Braunfeld v. Brown.  There, Pennsylvania required Sunday closings of certain businesses in order to establish “a day of community tranquility, respite and recreation, . . . when the atmosphere is one of calm and relaxation rather than one of commercialism.”  Orthodox Jewish merchants whose religion required them to close shop on Saturdays argued that this benefit for others imposed an impermissible burden on their own religious exercise, because it put them “at a serious economic disadvantage if they continue to adhere to their Sabbath,” i.e., if they closed for both weekend days rather than only for one.

The Court did not conclude that the indirect nature of the burden on the Jewish merchants foreclosed any and all constitutional analysis—to the contrary, Chief Justice Warren’s plurality opinion stated that it would be a “gross oversimplification” to “hold unassailable all legislation regulating conduct which imposes solely an indirect burden on the observance of religion.”  (That sentence, in fact, became the germ of the Sherbert decision in an “indirect burden” case two years later.)  Even so, the Court sustained the Sunday closing law—and did so only after making a pointed distinction about the dramatic differences between such indirect burdens and the burdens imposed by laws that actually require persons to do things their religion prohibits or prohibit them from doing what their religion requires.  To sustain the latter sort of law, the Chief Justice explained, “results in the choice to the individual of either abandoning his religious principle or facing criminal prosecution.” In contrast,

this is not the case before us because the statute at bar does not make unlawful any religious practices of appellants; the Sunday law simply regulates a secular activity and, as applied to appellants, operates so as to make the practice of their religious beliefs more expensive. . . .  Fully recognizing that the alternatives open to appellants and others similarly situated—retaining their present occupations and incurring economic disadvantage or engaging in some other commercial activity which does not call for either Saturday or Sunday labor—may well result in some financial sacrifice in order to observe their religious beliefs, still the option is wholly different than when the legislation attempts to make a religious practice itself unlawful. . . .  To strike down, without the most critical scrutiny, legislation which imposes only an indirect burden on the exercise of religion, i.e., legislation which does not make unlawful the religious practice itself, would radically restrict the operating latitude of the legislature.

Although the indirect nature of the pressure in Braunfeld did not excuse the state altogether from offering a justification for denying a religious exemption—the Chief Justice proceeded to inquire how the state interest might be affected by the conferral of religious exemptions—it obviously had a significant impact on the balance the Court struck.  Indeed, two years later, in distinguishing the result in Braunfeld, the Court in Sherbert v. Verner specifically noted that the burden on religious practice in that earlier case was “less direct” than the burden imposed on Adele Sherbert by the state’s threat to cut off her unemployment benefits unless she accepted work that would violate her religious obligations.

Here, if federal law places any pressure on some employers to retain their employee health insurance plans, it does so in a way far more attenuated than even the Sunday closing law in Braunfeld.  Such pressure is ultimately the result of ordinary competitive dynamics that are, in turn, the product of an interplay of a vast network of federal laws and market forces—a far cry from the pressure placed on Adele Sherbert when South Carolina threatened to cut off her unemployment benefits.  As the Court noted in Cutter v. Wilkinson (with respect to the same exact legal test, but under the Religious Liberty and Institutionalized Persons Act, rather than RFRA), the function of the Sherbert test, and the statutes that incorporate it, is to “alleviate[] exceptional government-created burdens on private religious exercise.” 544 U.S. at 720 (emphasis added).  The sorts of competitive pressures potentially at issue here are hardly exceptional, let alone created solely by the government.

Third, wholly apart from the fact that any pressure here is very indirect, the fact that a complex array of laws, predilections, market forces, motivations and competitor practices might cause many or most large employers to retain their health insurance plans does not prove that the government has put substantial pressure on such employers to do so.  Most employers might simply conclude that sticking with the status quo is the option that is marginally less risky or more convenient, all things considered.  The law creates countless modest incentives and disincentives that affect most people’s behavior because the alternative becomes marginally less convenient or more costly.  But the fact that an incentive is commonly felt does not mean that it imposes substantial pressure on persons to alter their conduct, akin to the pressure that South Carolina placed upon Adele Sherbert in Sherbert v. Verner.  Compliance with the condition in Sherbert, which would have required her to violate a religious precept, was necessary in order for Ms. Sherbert to receive unemployment benefits, which offered her the means of basic subsistence; therefore “the pressure upon her to forego [her religious] practice [was] unmistakable.”  374 U.S. at 404.  Whether or not federal law imposes such substantial pressure on any particular employer not to drop its health care plan is not nearly so certain, since here the government has not “condition[ed] receipt of an important benefit upon conduct proscribed by a religious faith.”  Thomas, 450 U.S. at 717.

Which brings me, finally, to the fourth, and most important, point for purposes of the Hobby Lobby and Conestoga Wood cases that the Court is considering:  Even if it would be fair to assume that federal law imposes substantial pressure on at least some employers to retain their employee health insurance plans, that is not established in these two cases.  Because those cases come to the Court at a very preliminary stage, with the record consisting solely of the complaints, the Court has no way of knowing just what sort of pressure the age-old tax exclusion for employees, and/or other provisions of federal law, would have on the decision-making of the two companies in question. 

For example:  The Court will not know how many, if any, of the plaintiffs’ employees would benefit from purchasing insurance on the exchanges; how much the employers currently pay in premiums and administrative costs that they would save if they no longer offered insurance; how much in the way of wage increases they would have to offer some employees to compensate for any losses those employees would realize by shifting to exchange-based insurance; the elasticity of the labor market for the employees in question; how important it is for these employers to retain such well-compensated employees; and so on, and so on.

Under RFRA the burden is on the plaintiffs to demonstrate a substantial burden on their religious exercise.  The Hobby Lobby and Conestoga Wood plaintiffs have alleged in their complaints that an actual legal obligation to offer and pay for employee contraception coverage would conflict with their owners’ religious obligations—and they have alleged that the ACA and the HHS “Mandate” create such a legal obligation because § 4980H(a) “penalizes” them for failing to offer employee health insurance.  See, e.g., Conestoga Wood Complaint paragraphs 52-53; Hobby Lobby Complaint paragraphs 120, 144. 

I have tried to explain in these posts that that’s a misreading of the law:  There is no duty to offer health insurance; § 4980H(a) does not impose a penalty for failing to do so; and the § 4980H(a) assessment is far less than the costs to the employer of providing an employee insurance plan.

Still, it is conceivable that the entire array of pre- and post-ACA federal laws, most especially the tax exception offered to employees for their employers’ insurance premiums, might result in substantial indirect pressure on these employers to retain their health insurance plans for competitive reasons. 

Conestoga Wood, however, does not mention this possibility in its complaint, let alone allege facts that would show that it is substantially pressured by law to retain its plan. 

The Hobby Lobby complaint, by contrast, does include a very general allegation (paragraph 139) that “[t]he Mandate places Plaintiffs at a competitive disadvantage in its efforts to recruit and retain employees”; see also paragraph 150 (“The Mandate exposes Plaintiffs to substantial competitive disadvantages, in that they may no longer be permitted to offer health insurance.”).  But it does not allege specific facts—indeed, it includes no facts at all—that would explain why and in what way the tax exception for employees would cause Hobby Lobby to suffer a “substantial competitive disadvantage” if it were to abandon its health insurance plan.  Under the Twombly/Iqbal doctrine, conclusory statements on a necessary element of plaintiffs’ case are almost certainly not sufficient to withstand a motion to dismiss, let alone to establish a likelihood of success justifying a preliminary injunction.  Hobby Lobby “would need to allege more by way of factual content” to move its claim of substantial burden “‘across the line from conceivable to plausible.’”  Iqbal, 556 U.S. at 683 (quoting Twombly, 550 U.S. at 570).

Older Posts
Newer Posts