Joel Raden et al., Appellants, v  W7879, LLC, et al., Respondents.

James Taylor et al., Respondents, v  72A Realty Associates, L.P., Appellant, et al., Defendant.

Elizabeth Reich, et al., Appellants, v  Belnord Partners, LLC, et al., Respondents.

Argued January 7, 2020; decided April 2, 2020

Matter of Regina Metro. Co., LLC v New York State Div. of Hous. & Community Renewal, 164 AD3d 420, affirmed.

Raden v W 7879, LLC, 164 AD3d 440, affirmed.

Taylor v 72A Realty Assoc., L.P., 151 AD3d 95, modified.

Reich v Belnord Partners, LLC, 168 AD3d 482, affirmed.

In our tripartite form of government, the legislature determines the public policy of this State, recalibrating rights and changing course when it deems such alteration appropriate as it grapples with enduring problems and rises to meet new challenges facing our communities. It is the distinct role of the courts to interpret the laws to give effect to legislative intent while safeguarding the constitutional rights of impacted individuals. We fulfill both core functions in these four appeals, which present a common issue under the Rent Stabilization Law (RSL): what is the proper method for calculating the recoverable rent overcharge for New York City apartments that were improperly removed from rent stabilization during receipt of J-51 benefits prior to our 2009 decision in Roberts v Tishman Speyer Props., L.P. (13 NY3d 270 [2009]).

As explained below, when leave was granted in these cases, the RSL mandated that, absent fraud, an overcharge was to be calculated by using the rent charged on the date four years prior to filing of the overcharge complaint (the lookback period) as the "base date rent," adding any legal increases applicable during the four-year lookback period and computing the difference between that legal regulated rent and the rent actually charged to determine if the tenant was overcharged during the recovery period. In such cases, consideration of rental history predating the four-year lookback and statute of limitations period was prohibited. While the appeals to this Court were pending,{**35 NY3d at 349} the legislature—as is its prerogative—enacted the Housing Stability and Tenant Protection Act of 2019 (HSTPA), making sweeping changes to the RSL, the majority of which are not at issue in these appeals. As relevant here, part F of the HSTPA includes amendments that, among other things, extend the statute of limitations, alter the method for determining legal regulated rent for overcharge purposes and substantially expand the nature and scope of owner liability in rent overcharge cases (see L 2019, ch 36, § 1, part F). The tenants in these cases urge us to apply the new overcharge calculation provisions to these appeals that were pending at the time of the HSTPA's enactment, some of which seek recovery of overcharges incurred more than a decade before the new legislation.

The validity of part F is not in question here—but significant issues are raised concerning whether the presumption against retroactive application of statutes has been rebutted and, if so, whether application of certain amendments relating to overcharge calculation in part F to these appeals involving conduct that occurred years prior to its enactment comports with fundamental notions of substantial justice embodied in the Due Process Clause. [*2]Retroactive application of the overcharge calculation amendments would create or considerably enlarge owners' financial liability for conduct that occurred, in some cases, many years or even decades before the HSTPA was enacted and for which the prior statutory scheme conferred on owners clear repose. Because such application of these amendments to past conduct would not comport with our retroactivity jurisprudence or the requirements of due process, we resolve these claims pursuant to the law in effect when the purported overcharges occurred. Notwithstanding the hyperbole employed by our dissenting colleagues, our analysis of the narrow legal issue presented by application of the overcharge calculation amendments to these appeals turns entirely on conventional and time-honored principles of judicial review. "We are, of course, mindful . . . of the responsibility . . . to defer to the Legislature in matters of policymaking," but it is the role of the judicial branch "to interpret and safeguard constitutional rights and review challenged acts of our co-equal branches of government—not in order to make policy but in order to assure the protection of constitutional rights" (Campaign for Fiscal Equity v State of New York, 100 NY2d 893, 925, 931 [2003]). As to the{**35 NY3d at 350} HSTPA, today we fulfill this quintessential judicial function in holding that a limited suite of enforcement provisions may not be applied retroactively and opine in no way on the vast majority of that legislation or its prospective application.

These rent overcharge cases arose in the wake of our 2009 decision in Roberts, interpreting RSL provisions relating to New York City's J-51 program, which offered tax benefits to building owners who made capital improvements to their residential properties. Buildings electing to receive J-51 benefits become subject to the rent stabilization scheme (RSL [Administrative Code of City of NY] § 11-243 [b], [i] [1]; [t]). From 1993 until the enactment of the HSTPA in 2019, the RSL contained "luxury deregulation" provisions, permitting an owner of a stabilized unit to deregulate if the rent exceeded a statutory threshold and (1) the tenant vacated or (2) the tenants' combined income exceeded a statutory threshold (former RSL §§ 26-504.1, 26-504.2). As early as 1996, first in an opinion letter and later promulgated as an agency regulation, the Division of Housing and Community Renewal (DHCR)^[FN1] took the position that statutory language precluding luxury deregulation of apartments during receipt of J-51 benefits did not apply to buildings that were already subject to the RSL prior to receipt of those benefits (see Roberts, 13 NY3d at 281-282; former Rent Stabilization Code [RSC] [9 NYCRR] § 2520.11 [r] [5]; [s] [2]). In Roberts, this Court rejected DHCR's long-standing statutory interpretation and concluded that luxury deregulation was unavailable in any building during receipt of J-51 benefits (13 NY3d at 285-287). In 2011, the Appellate Division held that Roberts applied retroactively (Gersten v 56 7th Ave. LLC, 88 AD3d 189, 198 [1st Dept 2011], appeal withdrawn 18 NY3d 954 [2012]).

Each of these cases involves an apartment that was treated as deregulated consistent with then-prevailing DHCR regulations and guidance before this Court rejected that guidance in Roberts. Indeed, the tenants took occupancy years prior to Roberts following a deregulation later revealed by that decision to have been improper, believing they were renting non-stabilized apartments at market rents. None of these tenants promptly challenged the deregulated status of their apartments and years—in some cases, over a decade—passed during which {**35 NY3d at 351}the tenants and their landlords renewed and renegotiated free-market leases.^[FN2] After we decided Roberts, these tenants commenced overcharge claims under the RSL. In Regina Metro., the tenants filed an administrative complaint with DHCR and, in the remaining three cases, the tenants commenced actions in Supreme Court. The central issue below in each of these cases—sent to this Court by leave of the Appellate Division before enactment of the HSTPA—was how to calculate [*3]the "legal regulated rent" in order to determine whether a recoverable overcharge occurred and its amount.^[FN3] Before we address the tenants' request that we resolve these appeals under the new law, we must determine the parties' rights under the statutory scheme in effect when the overcharges occurred.

In an overcharge claim, the tenant seeks monetary damages for excessive rent paid during the recovery period.^[FN4] The method {**35 NY3d at 352}for calculating the amount of recoverable damages—i.e., the overcharge—is governed by the RSL. We therefore examine the text of the relevant statutes, as the best indicator of legislative intent (Majewski v Broadalbin-Perth Cent. School Dist., 91 NY2d 577, 583 [1998]), mindful that legislative history may also be considered as an aid to interpretation (Altman v 285 W. Fourth LLC, 31 NY3d 178, 185 [2018]; see Riley v County of Broome, 95 NY2d 455, 463-464 [2000]). When a statute is part of a broader legislative scheme, we construe its language "in context and in a manner that harmonizes the related provisions and renders them compatible" (Matter of M.B., 6 NY3d 437, 447 [2006] [internal punctuation and citation omitted]).

The rules governing calculation of an overcharge are found in the provisions of the RSL addressing enforcement and the statute of limitations for overcharge claims (RSL § 26-516; CPLR 213-a). Before the enactment of the HSTPA, overcharge claims were subject to a four-year statute of limitations that precluded the recovery of overcharges incurred more than four years preceding the imposition of a claim (former RSL § 26-516 [a] [2]; former CPLR 213-a; see Conason v Megan Holding, LLC, 25 NY3d 1 [2015]). The statutes further directed that "no determination of an overcharge and no award or calculation of an award of the amount of an overcharge may be based [*4]upon an overcharge having occurred more than four years before" {**35 NY3d at 353}initiation of the claim (former RSL § 26-516 [a] [2]; see former CPLR 213-a).^[FN5]

A provision added as part of the Rent Regulation Reform Act of 1997 (1997 RRRA) expressly "preclude[d] examination of the rental history of the housing accommodation prior to the four-year period preceding" commencement of the overcharge action (former RSL § 26-516 [a] [2], as amended by L 1997, ch 116, § 33; see former CPLR 213-a, as amended by L 1997, ch 116, § 34)—language that "clarified and reinforced the four-year statute of limitations" (Thornton v Baron, 5 NY3d 175, 180 [2005]). This categorical temporal limitation on reviewable records—the "lookback" rule—was complemented by a record retention provision directing that certain owners "shall not be required to maintain or produce any records relating to rentals of such accommodation for more than four years prior to the most recent registration or annual statement for such accommodation" (former RSL § 26-516 [g]; see RSC § 2523.7 [b] ["An owner shall not be required to produce any rent records in connection with (overcharge) proceedings . . . relating to a period that is prior to the base date"]). The record retention provision permitted owners to dispose of records outside the four-year period (former RSL § 26-516 [g]; see Matter of Cintron v Calogero, 15 NY3d 347, 354 [2010]; Thornton, 5 NY3d at 181), further evincing the legislature's intent that records predating the recovery period not be used to calculate overcharges. Together, the statute of limitations, lookback provision and record retention rules formed an integrated scheme for calculating overcharges based on a closed universe of records pertaining only to the apartment's rental history in the four years preceding the filing of the complaint.

Consistent with the lookback rule, the enforcement provisions provided that, except for certain claims filed shortly after initial registration of a unit, "the legal regulated rent for purposes of determining an overcharge, shall be the rent indicated in the annual registration statement filed four years prior to the most recent registration statement," i.e., the base date rent, plus "any subsequent lawful increases and adjustments" (former RSL § 26-516 [a] [i]). Owners of rent-stabilized {**35 NY3d at 354}apartments are generally required to file annual rent registration statements with DHCR (RSL § 26-517 [f]), and where registration statements were filed during the lookback period, the base date rent was discerned from those statements. But owners are no longer required to file such statements once the apartment has been deregulated. Thus, where the apartment had been deregulated more than four years prior to the filing of an overcharge complaint, and the tenant failed to promptly challenge the deregulated status of the apartment, there might be no rent registration on file for the base date or, indeed, any time within the four-year lookback period.

This scenario is addressed in DHCR's regulations, which harmonized RSL § 26-516 (a) (i) with the four-year lookback restriction. With exceptions not relevant here, the regulations provided that "[t]he legal regulated rent for purposes of determining an overcharge shall be deemed to be the rent charged on the base date, plus in each case any subsequent lawful increases and adjustments" (RSC § 2526.1 [a] [3] [i] [emphasis added]; see also id. § 2520.6 [e]). Under the pre-HSTPA law, the base date rent was therefore the rent actually charged on the base date—i.e., four years prior to the overcharge complaint—even if no registration statement had been filed reflecting that rent.

In a series of cases, we confirmed that reviewing rental history outside the four-year lookback period was inappropriate for purposes of calculating an overcharge, but we recognized a limited common-law exception to the otherwise-categorical evidentiary bar, permitting tenants to use such evidence only to prove that the owner engaged in a fraudulent scheme to deregulate the apartment. In Thornton, the owner engaged in an egregious, fraudulent scheme to remove apartments from stabilization by conspiring with tenants, who shared in the illegal profits, by falsely agreeing the apartment was not being used as a primary residence (and utilizing the courts as a tool to obtain false declarations to that effect) to rent at market rates and then sublease at even higher rates (5 NY3d at 178-179). For overcharge calculation purposes, the Court acknowledged the preclusive effect of the four-year lookback rule, deeming the last regulated rent charged before that period to be "of no relevance" (id. at 180). We held that the legal rent should be based on a "default formula," otherwise reserved for cases where there are no reliable rent records, setting the base date rent as "the lowest rent charged for a rent-stabilized apartment {**35 NY3d at 355}with the same number of rooms in the same building on the relevant base date" (id. at 179-181 and n 1).

We elaborated on this fraud exception to the lookback rule in Matter of Grimm v State of N.Y. Div. of Hous. & Community Renewal Off. of Rent Admin., holding that where a tenant had made a "colorable claim of fraud" by identifying "substantial indicia," i.e., "evidence," of "a landlord's fraudulent deregulation scheme to remove an apartment from the protections of rent stabilization," that apartment's "rental history may be examined for the limited purpose of determining whether a fraudulent scheme to destabilize the apartment tainted the reliability of the rent on the base date" (15 NY3d 358, 366-367 [2010]). Consistent with Thornton, we directed that, if review of the rental history revealed such a fraudulent scheme, the default formula should be used to calculate any resulting overcharge (id. at 367). We confirmed this procedure in Conason, where the owner created a fictitious tenant and fictitious renovation to justify a rent increase (25 NY3d at 9, 16-17). Our holding in Matter of Boyd v New York State Div. of Hous. & Community Renewal (23 NY3d 999 [2014]), rejecting a challenge to DHCR's use of the rent actually charged four years prior to filing of the claim to calculate an overcharge in the absence of fraud, provided further clarification that the four-year lookback rule generally precluded review of rental history outside that period.

The rule that emerges from our precedent is that, under the prior law, review of rental history outside the four-year lookback period was permitted only in the limited category of cases where the tenant produced evidence of a fraudulent scheme to deregulate and, even then, solely to ascertain whether fraud occurred—not to furnish evidence for calculation of the base date rent or permit recovery for years of overcharges barred by the statute of limitations (Grimm, 15 NY3d at 367).^[FN6] In fraud cases, this Court sanctioned use of the default formula to set the base date rent. Otherwise, for overcharge calculation{**35 NY3d at 356} purposes, the base date rent was the rent actually charged on the base date (four years prior to initiation of the claim) and overcharges were to be calculated by adding the rent increases legally available to the owner under the RSL during the four-year recovery period. Tenants were therefore entitled to damages reflecting only the increases collected during that period that exceeded legal limits.

In the wake of Roberts, courts and DHCR grappled with a surge of claims filed by tenants alleging overcharges arising from the improper deregulation of their apartments years (in some cases more than a decade) before—claims like those now before this Court. For example, the plaintiffs in Raden, who took occupancy of their apartment in 1995 at a market rent, commenced this action in 2010 seeking recovery of overcharges based on a reconstruction of the rent they should have been charged had the apartment never been deregulated. Likewise, in Taylor, similar relief was sought in an overcharge claim filed in 2014 brought by a tenant who took occupancy in 2000. In stark contrast to Thornton, Grimm and Conason, in which tenants came forward with evidence of fraud,^[FN7] in these Roberts cases, the owners removed apartments from stabilization consistent with agency guidance. Deregulation of the apartments during receipt of J-51 benefits was not based on a fraudulent misstatement of fact but on a misinterpretation of the law—significantly, one that DHCR itself adopted and included in its regulations. As we observed in Borden v 400 E. 55th St. Assoc., L.P., a finding of willfulness "is generally not applicable to cases arising in the aftermath of Roberts" (24 NY3d 382, 398 [2014]). Because conduct cannot be fraudulent without being [*6]willful, it follows that the fraud exception to the lookback rule is generally inapplicable to Roberts overcharge claims.^[FN8]

{**35 NY3d at 357}After Roberts there was understandable confusion regarding how the decision should be implemented, including whether Roberts should be given retroactive effect and, if so, how that should be accomplished. In overcharge cases where tenants had not challenged the status of their apartments within four years of deregulation, including these appeals, the improper deregulation predated the lookback period and, thus, the rent charged on the "base date" was a free market rent that had not been registered. Tenants who challenged an improper deregulation and initiated an overcharge claim within four years would be entitled to monetary damages encompassing the rent increase that occurred when the apartment was moved to the free market. But tenants who commenced a claim more than four years later and could not show fraud would be entitled, by virtue of the interrelated four-year statute of limitations and lookback rule, to recover only the increases added to the market base date rent that were over the legal limits during the recovery period. This rule was applied properly in many cases (see Reich, 168 AD3d 482; Raden, 164 AD3d at 441; Stulz v 305 Riverside Corp., 150 AD3d 558 [1st Dept 2017], lv denied 30 NY3d 909 [2018]; see also Todres v W7879, LLC, 137 AD3d 597 [1st Dept 2016], lv denied 28 NY3d 910 [2016]). Yet, in some Roberts cases, DHCR and the lower courts deviated from the four-year limitations period and lookback rule in the absence of fraud.

The decision in 72A Realty Assoc. v Lucas (101 AD3d 401, 402 [1st Dept 2012]), which preceded our analysis in Boyd, represents such a deviation. In Lucas, the Appellate Division held that the four-year lookback rule should not be applied, even though the court did not find a colorable claim of fraud, in part because the rent charged four years prior to the complaint was a free market rent following improper deregulation. Citing Lucas, DHCR (in Regina Metro.) and the Appellate Division (in Taylor) determined that, even in the absence of fraud, an overcharge in a Roberts case should not be calculated in accordance with the four-year lookback rule but, instead, by reconstructing what the legal regulated rent would have been on the base{**35 NY3d at 358} date if the apartment had not been improperly deregulated. DHCR and the Taylor court determined that this reconstruction should be conducted by identifying the last legal regulated rent before improper deregulation—even though the apartment was deregulated more than four years prior to imposition of the claim—and applying all permissible rent increases between the date of that regulated rent and the base date (Regina Metro., 164 AD3d at 422-423; Taylor, 151 AD3d at 105-106).

The reconstruction method, applied by DHCR in Regina Metro. and approved by the Taylor court, violated the pre-HSTPA law by requiring review of rental history outside the four-year limitations and lookback period in the absence of fraud.^[FN9] The tenants' theory that Thornton, Grimm and Conason preclude adoption of a market base date [*7]rent is mistaken. Although in those cases we characterized base date rents resulting from fraud as "illegal" or "unreliable," we never suggested that an alternative method of setting the base date rent could apply to a less blameworthy owner where not authorized by the statutory scheme. Indeed, use of the reconstruction method violated the legislative mandate that "no award or calculation of an award of the amount of an overcharge may be based upon an overcharge having occurred more than four years before" (former RSL § 26-516 [a] [2]; see former CPLR 213-a). Moreover, it utilized rental history in a manner that this Court refused to sanction even in fraud cases, in which we authorized consideration of rental history outside the lookback period only for the "limited purpose" of determining whether a fraudulent scheme existed (Grimm, 15 NY3d at 367).

We are also unpersuaded by the tenants' arguments that use of a default formula or the other alternative approaches to{**35 NY3d at 359} determining base date rent^[FN10] would comply with pre-HSTPA law if applied to these cases. Even if employed in a manner compatible with the lookback rule, nothing in the RSL indicates that such methods apply here. While the alternative methods proposed by the tenants are reflected in the regulations, they are available only "[w]here the rent charged on the base date cannot be established" (RSC § 2526.1 [a] [3] [ii])—a situation not present in any of these Roberts cases.^[FN11]

The tenants and DHCR urge several bases for creating an exception to the standard pre-HSTPA overcharge calculation method that would enable courts to use these alternative approaches, but their arguments do not withstand scrutiny. First, an exception predicated on the fact that the base date rent was higher than what would have been permitted under the RSL for a stabilized apartment would swallow the four-year lookback rule. In every overcharge case, the rent charged was, by definition, illegally inflated—otherwise there would be no overcharge. Prior to the HSTPA, nothing in the rent stabilization scheme suggested that where an unrecoverable overcharge occurred{**35 NY3d at 360} before the base date, thus resulting in a higher base date rent, the four-year lookback rule operated differently. To the contrary, the limitations provisions—in order to promote repose—precluded consideration of overcharges prior to the recovery period (former RSL § 26-516 [a] [2]; former CPLR 213-a), and it is clear from Boyd that use of a potentially inflated base date rent, flowing from an overcharge predating the limitations and lookback period, was proper in the absence of fraud. Likewise, no exception is justified by the fact that the inflated base date rent in Roberts cases resulted from improper deregulation, as opposed to an improperly high increase to a stabilized rent. The RSL makes no such distinction, and there is no indication that, under the pre-HSTPA law, an overcharge resulting from improper (but non-fraudulent) luxury deregulation warranted anything but the application of the standard lookback provisions.

Nor is it necessary to recognize an additional common-law exception that would create or increase the amount of overcharge damages in order to give proper effect to Roberts. Civil liability is always bounded by the public policy of repose embodied in statutes of limitations (see Ajdler v Province of Mendoza, 33 NY3d 120, 130 n 6 [2019] ["(T)he (s)tatute of (l)imitations . . . expresses a societal interest or public policy of giving repose to human affairs"], quoting John J. Kassner & Co. v City of New York, 46 NY2d 544, 550 [1979]). Overcharge liability under the RSL is no different. That Roberts revealed particular conduct to be illegal does not mean that tenants must be able to recover a certain measure of monetary damages for associated rent increases despite their failure to seek recovery within the limitations and lookback periods. Critically, our decision in Roberts has led to the return of many apartments to the rent stabilization scheme, including those at issue in these appeals; one amicus estimates the number of Roberts apartments at upwards of 50,000. While the statute of limitations and lookback period preclude tenants in those apartments from recovering certain damages they could have recovered if their claims had been initiated earlier, as a result of Roberts they may now enjoy rent stabilization protection.

Indeed, in Taylor, regardless of any entitlement to monetary damages, the tenants' request for a declaration that the apartment was rent-stabilized at the time of their complaint was properly granted. RSL § 26-504 (c) provides pathways by which an apartment in a building receiving J-51 benefits may be{**35 NY3d at 361} deregulated at the conclusion of the benefit period.^[FN12] Particularly relevant here, section 26-504 (c) states that

"if such dwelling unit would have been subject to [the RSL or the Emergency Tenant Protection Act (ETPA)] in the absence of [J-51 benefits or certain other programs], such dwelling unit shall, upon the expiration of such benefits, continue to be subject to [the RSL or ETPA] to the same extent and in the same manner as if [section 26-504 (c)] had never applied thereto."

Thus, in buildings affected by Roberts, all of which were subject to the RSL regardless of J-51 benefits, apartments revert to their original rent-stabilized status after expiration of J-51 benefits.^[FN13]

We therefore decline to create a new exception to the lookback rule and instead clarify that, under pre-HSTPA law, the four-year lookback rule and standard method of calculating legal regulated rent govern in Roberts overcharge cases, absent fraud. Applying the correct interpretation of the pre-HSTPA law to the present cases, in Regina Metro. the Appellate Division properly annulled DHCR's overcharge determination, which violated the lookback rule by relying on a reconstructed{**35 NY3d at 362} rent, despite finding that the overcharge was not willful (and there was no colorable fraud claim). In Raden, the de minimis overcharge was properly calculated using the standard method, accepting the rent charged on the base date as the base date rent and adding legal increases. In Reich, the complaint was properly dismissed based on the tenants' failure to allege a colorable claim of fraud and the absence of allegations indicating that, applying the standard overcharge calculation method, there was an overcharge during the recovery period. And [*9]in Taylor, modification of the Appellate Division order is necessary to grant summary judgment dismissing the overcharge claim based on the owner's unrebutted evidence that there were no overcharges using the standard calculation method.

Normally, our analysis would end here. But the HSTPA, enacted in June 2019 and consisting of 15 parts, substantially revised New York's rent stabilization scheme by, among other things, eliminating luxury deregulation, amending mechanisms for rent increases and providing for the expansion of regulation to new geographic areas (L 2019, ch 36). Here, although the alleged overcharges occurred years in the past, well before the HSTPA was enacted, the tenants ask us to apply certain amendments revising the enforcement provisions of the RSL with respect to overcharge claims, all contained in part F of the legislation.^[FN14] Although we generally do not review issues raised for the first time on appeal, we may consider the applicability{**35 NY3d at 363} of this new legislation enacted while these appeals were pending in this Court, "which could not have been raised below as those proceedings predated the amendment" (Matter of Gleason [Michael Vee, Ltd.], 96 NY2d 117, 121 n [2001]).^[FN15] In the context of legislation as significant as the HSTPA, the question we address here is relatively narrow—we have no occasion to address the prospective application of any portion of the HSTPA, including part F. We address the new legislation only to determine whether [*10]certain part F amendments discussed below must be applied retroactively to past conduct—and therefore govern these appeals, as urged by the tenants. We conclude that the overcharge calculation amendments cannot be applied retroactively to overcharges that occurred prior to their enactment.

Part F extended the four-year limitations period for overcharge claims to six years, provided that an overcharge complaint "may be filed . . . at any time" and eliminated the provision—present, in substance, since 1983—stating that "no determination of an overcharge and no award or calculation of an award of the amount of an overcharge may be based upon an overcharge having occurred more than four years before the{**35 NY3d at 364} complaint is filed" (RSL § 26-516 [a] [2]; see CPLR 213-a). It also entirely abolished the lookback rule in favor of new requirements: the base date rent is no longer defined as the rent charged or reflected in a registration statement on the base date but that reflected in the "most recent reliable" registration statement filed six "or more" years before the most recent registration (RSL § 26-516 [a] [i]). Examination of rent history that predates the period covered by the former lookback rule is no longer precluded. Instead, DHCR and courts are now required to "consider all available rent history which is reasonably necessary" to investigate overcharge claims and determine legal regulated rent, regardless of the vintage of that history and including records kept by owners, tenants and agencies (id. § 26-516 [a] [i]; [h]). Part F likewise lengthened the four-year record retention period to six years and provides that an owner's "election not to maintain records" does not limit the authority of DHCR or a court to examine the rental history further (id. § 26-516 [g]). Whereas the RSL previously provided for only two years of treble damages for willful overcharges, treble damages are now recoverable for the entire six-year limitations period (id. § 26-516 [a] [2]).^[FN16]

The tenants argue that these amendments should be applied to these appeals based on the provision stating that part F "shall take effect immediately and shall apply to any claims pending or filed on and after such date" (see L 2019, ch 36, § 1, part F, § 7). The owners argue that the effective date language does not evince a clear legislative intent to apply the new overcharge calculation provisions retroactively, particularly to cases no longer pending in DHCR or the trial court and further contend, in any event, that retroactive application of the new overcharge calculation methodology to these appeals would violate due process protections in the State and Federal Constitutions. We must first assess whether applying these amendments to overcharges that occurred before the HSTPA's{**35 NY3d at 365} enactment truly implicates the concerns historically associated with retroactive application of new legislation.

In Landgraf v USI Film Products, the Supreme Court articulated a contemporary framework for analyzing retroactivity—adopted by this Court—which recognized that application of a new statute to conduct that has already occurred may, but does not necessarily, have "retroactive" effect upsetting reliance interests and triggering fundamental concerns about fairness (511 US 244 [1994]; see also American Economy Ins. Co. v State of New York, 30 NY3d 136, 149 [2017], cert denied 584 US —, 138 S Ct 2601 [2018]). Landgraf harmonized the "apparent tension" between the presumption against retroactive application of statutes and statutory construction canons applied in prior cases to discern a statute's temporal scope, which concerned statutes with no truly retroactive effect (511 US at 263-280; see e.g. Bradley v School Bd. of Richmond, 416 US 696 [1974] [holding a newly enacted statute authorizing the award of a reasonable attorney's fee to a prevailing party in a school desegregation case could be relied on in a pending action to support a claim for such a fee for legal services rendered before the statute was enacted]; Thorpe v Housing Authority of Durham, 393 US 268, 278-279 [1969] [holding that a new agency policy imposing "a very simple notification procedure" that a housing authority had to follow prior to evicting a tenant, which did not alter the lease terms or take away the housing authority's legal ability to evict, was applicable to an eviction proceeding commenced before the policy was issued but not yet completed]; see also Landgraf, 511 US at 285 n 37 [likewise limiting the continued utility of the tenet that new "remedial" statutes apply presumptively to pending cases]).

A statute has retroactive effect if "it would impair rights a party possessed when he acted, increase a party's liability for past conduct, or impose new duties with respect to transactions already completed," thus impacting "substantive" rights (Landgraf, 511 US at 278-280; see also American Economy, 30 NY3d at 147). On the other hand, a statute that affects only "the propriety of prospective relief" or the nonsubstantive provisions governing the [*11]procedure for adjudication of a claim going forward has no potentially problematic retroactive effect even when the liability arises from past conduct (Landgraf, 511 US at 273; see e.g. Ex parte Collett, 337 US 55, 71 [1949]{**35 NY3d at 366} [transfer of a civil action]).^[FN17] For example, in Matter of Raynor v Landmark Chrysler, this Court held that a legislative amendment revising the "time and manner" of insurers' payments for future workers' compensation awards arising from prior injuries did not have retroactive effects because "the statute neither altered the carrier's preexisting liability nor imposed a wholly unexpected new procedure" (18 NY3d 48, 57 [2011]).^[FN18]

Landgraf illustrates this distinction. There, the Supreme Court addressed whether 1991 amendments to title VII of the Civil Rights Act of 1964 applied to pending litigation. Prior to 1991, the primary monetary relief available under title VII was back pay for lost wages, recoverable only if unlawful discrimination had a concrete effect on the plaintiff's employment (511 US at 252-254). The 1991 act "significantly expand[ed] the monetary relief potentially available to plaintiffs"—and "allow[ed] monetary relief for some [cases] that would not previously have justified any relief under Title VII"—by providing for compensatory damages (including for future pecuniary losses and nonpecuniary losses) and punitive damages and making monetary damages recoverable even absent a concrete effect on employment, as well as created a right to a jury trial in certain damages cases (id.). The Court deemed the jury trial provision purely procedural with no retroactive effect if applied to cases that had been commenced but had not yet proceeded to trial before the statute was enacted{**35 NY3d at 367} (id. at 280-281 and n 34). On the other hand, the punitive damages provision was "clearly" retroactive if applied to conduct occurring before the statute's enactment, as it reflected a punishment for past acts (id. at 281). The compensatory damages provision—which was "quintessentially backward looking"—also would have had a retroactive effect because, in cases where money damages were previously unrecoverable, it would "attach an important new legal burden" and could "be seen as creating a new cause of action" (id. at 282-283). The Court noted that, even in cases where monetary damages were previously available, the new provision "resemble[d] a statute increasing the amount of damages available under a preestablished cause of action" that would, if applied to pending cases, "undoubtedly impose . . . a 'new disability' in respect to past events," explaining that the "extent of a party's liability, in the civil context as well as the criminal, is an important legal consequence" in determining retroactivity (id. at 283-284 [citation omitted]).

Here, if applied to past conduct, the amendments to the statute of limitations, overcharge calculation and damages provisions in part F of the HSTPA would impose new liability and thus have a "retroactive effect"—altering substantive rights in multiple ways. The statute of limitations with respect to overcharge claims has been treated as running backward from the date of initiation of the claim, previously permitting recovery of overcharges occurring only in each of the four years preceding the complaint. Thus, the relevant illegal conduct for which a tenant can recover is the overcharge committed in any given year during the recovery period. Expansion of the limitations period from four to six years clearly has a retroactive effect because it permits recovery for nonfraudulent conduct occurring during an additional two years preceding the former recovery period—conduct that was beyond challenge [*12]under the prior law. Likewise, the imposition of treble damages for four additional years of overcharges—conduct not previously subject to treble damages—clearly increases the scope of liability for past wrongs if applied retroactively, as the Supreme Court indicated in Landgraf (id. at 281).

Critically, for purposes of calculating the amount owed for any overcharge, part F now renders reviewable rent increases that were shielded by the prior lookback rule, permitting reconstruction of the legal regulated rent based on any relevant records in the apartment's entire rental history. Although{**35 NY3d at 368} the tenant can directly recover only for overcharges occurring during the six years preceding the complaint, the damages calculations for those years may now effectively incorporate conduct—illegal increases—preceding that period and occurring at any point in the rental history. This amendment is not merely, as the dissent contends, a procedural change regarding what evidence can be considered (dissenting op at —); it expands the scope of owner liability significantly based on conduct that was inoculated by the old law.^[FN19] In the same way that the compensatory damages provision in Landgraf would have provided monetary relief for conduct that, while illegal, previously did not provide a right to such relief, the effect here would be to permit recovery, previously barred by the lookback rule and limitations period, for past conduct that violated the RSL. Even if the amendments could be viewed in some cases as merely increasing damages for conduct that already gave rise to monetary relief, the dissent is wrong that such "tinker[ing] with the recoverable amount" has no retroactive effect (dissenting op at —, —). Under Landgraf, statutes that expand "[t]he extent of a party's liability" under the same cause of action have retroactive effect (511 US at 283-284 [observing that in no case "in which Congress had not clearly spoken, ha(d) (the Court) read a statute substantially increasing the monetary liability of a private party to apply to conduct occurring before the statute's enactment"]).

This retroactive effect becomes even more pronounced when considered in tandem with the HSTPA amendments to the record retention requirements. Those amendments expand the retention period by two years and, although the provision still nominally permits an owner to destroy some records—now after six years—the new law states that "an owner's election not to maintain records shall not limit the authority of [DHCR] and the courts to examine the rental history and determine legal regulated rents" (RSL § 26-516 [g]). Thus, the HSTPA effectively provides that an owner can be penalized indirectly for a disposal of records that was legal under the prior law but{**35 NY3d at 369} will now hinder the owner's ability to establish the legality of (and non-willfulness of any illegal) rent increases outside the lookback period, which—under the new legislation—impact recovery even in the absence of fraud.^[FN20]

Retroactive application of the overcharge calculation provisions in part F implicates all three Landgraf retroactivity criteria by impairing rights owners possessed in the past, increasing their liability for past conduct and imposing new duties with respect to transactions already completed. This is true even though rent stabilization is a highly regulated area. As we explained in American Economy when addressing the rights of employers' insurers under another highly regulated regime—workers' compensation—this is an area designed with "flexibility," in which "[t]he allocation of economic benefits and burdens has always been subject to adjustment" (30 NY3d at 148-149, quoting [*13]Becker v Huss Co., 43 NY2d 527, 541 [1978]). "The Constitution merely mandates that a landlord earn a reasonable return," and no party doing business in a regulated environment like the New York City rental market can expect the RSL to remain static, as we have repeatedly made clear in cases challenging prospective legislation altering the formula for rent increases under prior schemes (see I. L. F. Y. Co. v City Rent & Rehabilitation Admin., 11 NY2d 480, 492 [1962]; Bucho Holding Co. v Temporary State Hous. Rent Commn., 11 NY2d 469 [1962]). But applying these amendments to past conduct is not related to legislative decisions about proper division of economic burdens going forward, and it does not simply upset expectations about the continuing future availability of a favorable regulatory mechanism. Rather, by increasing overcharge exposure relating to owners' past acts, retroactive application of the provisions would undermine considerable reliance{**35 NY3d at 370} interests concerning income owners already derived from rents collected on real property years—if not decades—before.

Because the overcharge calculation provisions, if applied to past conduct, would impact substantive rights and have retroactive effect, the presumption against retroactivity is triggered. As opposed to a decisional change in the common law—which typically but not invariably applies "to all cases still in the normal litigating process" (Gurnee v Aetna Life & Cas. Co., 55 NY2d 184, 190-191 [1982] [citation omitted] [permitting retroactive application of interpretation of insurance law "to all claims not barred by the Statute of Limitations"])—generally, a statute is presumed to apply only prospectively (Majewski, 91 NY2d at 584). Retroactive legislation is viewed with "great suspicion" (Matter of Chrysler Props. v Morris, 23 NY2d 515, 521 [1969]). This "deeply rooted" presumption against retroactivity is based on "[e]lementary considerations of fairness [that] dictate that individuals should have an opportunity to know what the law is and to conform their conduct accordingly" (Landgraf, 511 US at 265). As the Supreme Court has cautioned, careful consideration of retroactive statutes is warranted because "[t]he Legislature's unmatched powers allow it to sweep away settled expectations suddenly and without individualized consideration" and "[i]ts responsivity to political pressures poses a risk that it may be tempted to use retroactive legislation as a means of retribution against unpopular groups or individuals" (id. at 266).

In light of these concerns, "[i]t takes a clear expression of the legislative purpose . . . to justify a retroactive application" of a statute (Gleason v Gleason, 26 NY2d 28, 36 [1970] [internal quotation marks and citation omitted]), which "assures that [the legislative body] itself has affirmatively considered the potential unfairness of retroactive application and determined that it is an acceptable price to pay for the countervailing benefits" (Landgraf, 511 US at 272-273). The ultimate question here, therefore, is one of statutory interpretation: whether the legislature has expressed a sufficiently clear intent to apply the overcharge calculation amendments retroactively to these pending appeals. There is certainly no requirement that particular words be used—and, in some instances, retroactive intent can be discerned from the nature of the legislation (see e.g. Eastern Enterprises v Apfel, 524 US 498 [1998]; Usery v Turner Elkhorn Mining Co., 428 US 1 [1976]). But the expression of intent must be sufficient to show that the legislature contemplated{**35 NY3d at 371} the retroactive impact on substantive rights and intended that extraordinary result. Even within the same legislation, language may be sufficiently clear to effectuate application of some amendments to cases arising from past conduct but not others with more severe retroactive effect (see Landgraf, 511 US at 280-281; Matter of Beary v City of Rye, 44 NY2d 398, 410-411 [1978]).

If retroactive application would not only impose new liability on past conduct but also revive claims that were time-barred at the time of the new legislation, we require an even clearer expression of legislative intent than that needed to effect other retroactive statutes—the statute's text must unequivocally convey the aim of reviving claims. For nearly a century, this Court has recognized that "[r]evival is an extreme exercise of legislative power. The will to work it is not deduced from words of doubtful meaning. Uncertainties are resolved against consequences so drastic" (Hopkins v Lincoln Trust Co., 233 NY 213, 215 [1922, Cardozo, J.]). Indeed, it is a bedrock rule of law that, absent an unambiguous statement of legislative intent, statutes that revive time-barred claims if applied retroactively will not be construed to have that effect (see e.g. Matter of Thomas v Bethlehem Steel Corp., 63 NY2d 150, 155 [1984]; Beary, 44 NY2d at 412-413). For example, in 35 Park Ave. Corp. v Campagna, plaintiff contended that a newly enacted statute permitting a court to grant relief from an unconscionable lease or clause—which the legislature deemed "applicable to all leases, regardless of when executed"—revived a time-barred claim to rescind a lease (48 NY2d 813, 814-815 [1979]). Citing the need for clear and unequivocal language "to effect so drastic a consequence," the Court reasoned that the language rendering the statute "applicable to all leases" was "ambiguous," failing to convey a sufficiently clear intention to resurrect time-barred claims (id. at 815).

When the legislature has intended to revive time-barred claims, it has typically said so unambiguously, providing a limited window when stale claims may be pursued. For example, Jimmy Nolan's Law, which we [*14]addressed in Matter of World Trade Ctr. Lower Manhattan Disaster Site Litig. (30 NY3d 377 [2017]), expressly "revived" certain time-barred claims related to World Trade Center cleanup and rescue work, permitting suit during a discrete one-year window period (see General Municipal Law § 50-i [4] [a], as added by L 2009, ch 440, § 2). Similar unequivocal "revival" language accompanied {**35 NY3d at 372}by a limited period for commencement of time-barred claims appears in the statute reviving toxic tort cases, including those arising from exposure to the drug diethylstilbestrol ingested by pregnant women (L 1986, ch 682, § 4), addressed in Hymowitz v Eli Lilly & Co. (73 NY2d 487 [1989]). The legislature has historically acted with deliberation and clarity when upsetting the strong public policy favoring finality, predictability, fairness and repose served by statutes of limitations.^[FN21]

The "claims pending" language in the part F effective date provision is insufficient to indicate that the legislature intended retroactive application in a manner that revives time-barred claims, such as by extending the statute of limitations to permit recovery of two annual overcharge claims that were time-barred under the prior law. This language bears no resemblance to the express claim revival language in the statutes addressed by World Trade Ctr. and Hymowitz—yet the claim revival effect if the relevant amendments to the HSTPA were to be applied retroactively is substantially more far-reaching than that of the orderly and even-handed claim revival method used in those statutes, which created a narrow window for commencement of time-barred suits. If applied to past conduct, the relevant HSTPA amendments would not only revive claims for two additional years but, by changing the overcharge calculation methodology to enable review of any illegal rent increase in the history of the apartment, would also substantially alter the nature of the liability by resurrecting nonfraudulent overcharges that initially occurred more than four years prior to the complaint but continue to impact the calculation of the current rent. Just as the statutory language in 35 Park Ave. Corp., rendering the new legislation "applicable to all leases, regardless of when executed" (48 NY2d at 814-815), fell short of our standard, here the generic reference to "any claims pending" upon enactment does not provide the requisite textual assurance that the legislature considered the significant impact of reviving barred claims, upsetting the strong public policy favoring repose, and that it desired that result.

This does not entirely resolve the statutory interpretation question, however, because as we have explained, retroactive{**35 NY3d at 373} application of part F would have significant impacts beyond claim revival, specifically on the scope and nature of damages recoverable with respect to timely claims. While the presumption against claim revival effect may only be overcome by the legislature's unequivocal textual expression that the statute was intended not only to apply to past conduct, but specifically to revive time-barred claims (see 35 Park Ave. Corp., 48 NY2d at 815), the general presumption against retroactive effect may be overcome by either an express prescription of the statute's temporal reach or a less explicit but "comparably firm conclusion"—applying "normal rules of construction"—of legislative intent to apply the enactment to conduct that occurred previously (Fernandez-Vargas v Gonzales, 548 US 30, 37 [2006] [citation omitted]; see also Majewski, 91 NY2d at 584). Although the HSTPA part F effective date provision does not express an intent to revive time-barred claims under our heightened claim revival standard, read in the specific context of this legislation, the "claims pending" language is sufficiently clear to evince legislative intent to apply the amendments to at least some timely overcharge claims that were commenced prior to enactment.

Each of the HSTPA's 15 parts contains its own effective date provision, indicating the legislature considered the issue of temporal scope for each. The legislation is almost entirely forward-looking—only part F's effective date provision contains language referring to prior claims. In contrast, many of the HSTPA's other effective date provisions, such as that applicable to the amendments eliminating vacancy and longevity bonuses, state only that the parts of the legislation to which they apply "shall take effect immediately" (see L 2019, ch 36, § 1, part A, § 7; part B, § 8; part C, § 5; part D, § 8; part G, § 7; part J, § 2; part L, § 3), in some cases indicating when the amendments contained therein expire (id. part E, § 3; part H, § 5; part I, § 5; part K, § 18). Others expressly provide that the relevant part applies prospectively only, such as by indicating that it takes effect immediately but applies to actions "commenced on or after such effective date" or that certain amendments take effect at some point in the future, such as "on the thirtieth day after this act shall have become a law" (id. part M, § 29; see also id. part N, § 2 [part N "shall take effect immediately and shall only apply to plans (for conversion of an apartment to a condominium or cooperative) [*15]submitted . . . after the effective date"]; part O, § 14 [part O "shall take effect {**35 NY3d at 374}on the thirtieth day after it shall have become a law"]). Therefore, this is not a case where the legislature passed comprehensive legislation, including general "claims pending" language, without differentiating between the parts it intended to apply retroactively and those that could reasonably be given only prospective effect. Moreover, part F relates almost entirely to the calculation of overcharge claims, and any such claim that was pending at the time the HSTPA was enacted necessarily involved conduct that occurred prior to the statute's enactment.

Read in context, and because some of the part F provisions have effects beyond reviving time-barred claims, the "claims pending" language must be construed as evincing a retroactive intent.^[FN22] At the very least, "claims pending" indicates the part F provisions were intended to apply to overcharge claims where the calculation issue remained unresolved as of the June 2019 effective date. Indeed, in Landgraf, the Supreme Court indicated that similar language referencing "pending" cases would have been sufficient in that case to reflect a retroactive intent (see Landgraf, 511 US at 259-260 [referencing language in a prior version of the statute stating the provisions "shall apply to all proceedings pending on or commenced after the date of enactment"]). Therefore, although there was no clear directive to revive time-barred claims, we conclude that the legislature evinced a sufficiently clear intent to apply part F to timely pending claims, such as Regina Metro. and Taylor, where the overcharge calculation issue was unresolved at the time the HSTPA was enacted.^[FN23] It is therefore necessary to reach the constitutional challenge.{**35 NY3d at 375}

To comport with the requirements of due process, retroactive application of a newly enacted provision must be supported by "a legitimate legislative purpose furthered by rational means" (American Economy, 30 NY3d at 157-158, citing General Motors Corp. v Romein, 503 US 181, 191 [1992]). Of course, as with prospective elements of legislation, legislative direction concerning the scope of a statute carries a presumption of constitutionality, and the party challenging that direction bears the burden of showing the absence of a rational basis justifying retroactive application of the statute (Turner Elkhorn, 428 US at 15). Nevertheless, the Supreme Court has made clear that "retroactive legislation does have to meet a burden not faced by [purely prospective] legislation," which is satisfied when "the retroactive application of the legislation is itself justified by a rational legislative purpose" (Pension Benefit Guaranty Corporation v R. A. Gray & Co., 467 US 717, 730 [1984] [emphasis added]).

Because "[r]etroactive legislation presents problems of unfairness that are more serious than those posed by prospective legislation" (Romein, 503 US at 191), "the justifications for [prospective legislation] may not suffice for the [retroactive aspects]" (R. A. Gray & Co., 467 US at 730). We have suggested that, in order to comport with due process, there must be a "persuasive reason" for the "potentially harsh" impacts of retroactivity (Holly S. Clarendon Trust v State Tax Commn., 43 NY2d 933, 935 [1978]; see Chrysler Props., 23 NY2d at 522 [there was no "persuasive case" supporting retroactive application]). Our acknowledgement that retroactive legislation must be supported by a rational basis commensurate with the degree of retroactive effect does not represent a "bifurcation" between the rational basis analyses for prospective and retroactive legislation (see dissenting op at —). Consideration of the scope of legislation is critical to a rational basis analysis, regardless of whether it is solely prospective or also involves retroactive effects.

In tax cases, an area where retroactive application of statutes is more highly tolerated, if for a short time (James Sq. Assoc. LP v Mullen, 21 NY3d 233, 246 [2013], citing Matter of Replan Dev. v Department of Hous. Preserv. & Dev. of City of {**35 NY3d at 376}N.Y., 70 NY2d 451, 455 [1987], and Welch v Henry, 305 US 134, 146 [1938]), we have highlighted particular factors relevant to the due process analysis for retroactive legislation. In Replan, we explained that whether a retroactive statute comports with due process principles is a "question of degree" that turns on the length of the retroactivity period, the taxpayer's forewarning of a change in legislation as relevant to reliance interests and the public purpose for retroactive application (70 NY2d at 456). Our consideration of these factors—derived from Supreme Court precedent—"does not differ from the prohibition against arbitrary and irrational legislation that applies generally to enactments in the sphere of economic policy" (Caprio v New York State Dept. of Taxation & Fin., 25 NY3d 744, 752 [2015], quoting United States v Carlton, 512 US 26, 30 [1994]). Instead, in requiring that there be a non-arbitrary justification for retroactive application of a statute, the rational basis test incorporates the equitable considerations that Replan highlights more directly. Likewise, the Supreme Court has indicated that it applies the same due process analysis in the tax context that applies to any other economic legislation (Carlton, 512 US at 30), albeit recognizing that retroactivity is more tolerable in tax legislation (see United States v Darusmont, 449 US 292, 296-298 [1981]; Welch, 305 US at 146, 149-150).^[FN24]

In determining whether retroactive application of a statute is supported by a rational basis, the relationship between the length of the retroactivity period and its purpose is critical. Generally, there are two types of retroactive statutes that courts have found to be constitutional: those employing brief, defined periods that function in an administrative manner to assist in effectuating the legislation, and statutory retroactivity that—even if more substantial—is integral to the fundamental aim of the legislation. For example, in the first category, courts have rejected challenges to the legislative practice of{**35 NY3d at 377} incorporating a clear, limited retroactivity period intended to prevent parties from taking advantage of a lengthy legislative process to circumvent a statute. In R. A. Gray & Co., the Supreme Court rejected a due process challenge to retroactive application of a statute that required employers that withdrew from a multiemployer pension plan to pay a fixed debt to the plan and expressly extended that penalty to those who withdrew within the five months prior to enactment (467 US at 720, 725). After observing that Congress had been "quite explicit" that the statute was made retroactive in order to "prevent employers from taking advantage of a lengthy legislative process and withdrawing while Congress debated," the Court emphasized that the retroactivity period was limited in scope to achieve its aim, noting, "as the amendments progressed through the legislative process, Congress advanced the effective date chosen so that it would encompass only that retroactive time period that Congress believed would be necessary to accomplish its purposes" (467 US at 730-731).

Falling within the latter category—instances where retroactive application was central to the statutes' purpose—in Turner Elkhorn the Supreme Court upheld legislation requiring coal operators to compensate miners who had already left the industry for the disability caused by the latent effects of exposure to coal dust, resulting in black lung disease, or pneumoconiosis (428 US at 15, 18-20). The Supreme Court reasoned that the retroactive imposition of liability on the coal operator that previously employed the miner was "justified as a rational measure to spread the costs of the employees' disabilities to those who have profited from the fruits of their labor" (id. at 18; but see Eastern Enterprises v Apfel, 524 US 498 [1998] [deeming retroactive application of a coal miner health care benefit scheme, requiring participation by a company that ceased coal mining operations in 1965, unconstitutional]). Similarly, in American Economy—where we assumed without deciding that a statute closing a fund that previously benefitted workers' compensation insurance carriers liable for "reopened" claims had a retroactive impact—we concluded application of the statute to injuries incurred under workers' compensation insurance policies finalized prior to the effective date was necessary to achieve its purpose (30 NY3d at 158-159). There, the workers' compensation fund [*17]was closed to relieve the burden on employers supporting its costs, which had increased dramatically in the six years preceding the{**35 NY3d at 378} legislation due to skyrocketing medical costs and an unexpected surge in reopened cases (id. at 143). If that closure was not applied to claims arising from past injuries, the fund "would have incurred substantial new liabilities for many years, given the duration of many workers' compensation cases," and "the relief to businesses sought by the legislature would have been indefinitely delayed" (id. at 158). In cases where retroactivity is integral to full achievement of the fundamental purpose of the legislation, a rational basis for the retroactive effect may be readily identifiable.^[FN25]

On the other hand, even short periods of retroactivity will be invalidated absent the requisite rational basis. In Chrysler Props., we sustained a constitutional challenge to retroactive application of a statute providing New York City a new right to seek judicial review of adverse determinations of the State Tax Commission, despite the relatively brief, four-month period of retroactivity set forth in the effective-date provision (23 NY2d at 518-519). Because there previously was no right to challenge such determinations, we concluded that a taxpayer who was issued a refund order only one month before the new law was entitled to payment because the taxpayer "had obtained a sufficiently certain right to the money" and the legislature made the amendment retroactive "without any discernable reason" (id. at 517-519). Likewise, in James Sq., we invalidated the retroactive application of amendments to the Empire Zones Program Act that changed the criteria for receipt of tax benefits, noting businesses had no forewarning of the change, and that a 16-month period of retroactivity was excessive because businesses had "gained a reasonable expectation that they would secure repose in the existing tax scheme" (21 NY3d at 248-250 [internal quotation marks omitted]). We emphasized that retroactively denying tax credits did not further any aim of the statute—by spurring investment or preventing abuses of {**35 NY3d at 379}the program—but "simply punished . . . participants more harshly for behavior that already occurred and that they could not alter" (id. at 250). Unlike the statutes at issue in R. A. Gray & Co., Turner Elkhorn or American Economy, where the retroactive scope was directly related or integral to furtherance of the legislative goals, in Chrysler Props. and James Sq. we concluded retroactive application would be irrational given the extent of settled interests, degree of repose and lack of a permissible basis for unsettling those interests.

Those same concerns are amplified here. The HSTPA's overcharge calculation and treble damages provisions, if applied retroactively, would more severely impact substantive rights than the provision in James Sq., which involved a tax statute, an area where courts are generally more tolerant of retroactivity. Before the HSTPA, the combined effect of the statute of limitations and lookback rule provided owners substantial repose relating to rent increases collected more than four years prior to the filing of the complaint. In sharp contrast, the HSTPA amendments directing review of all available rental history to reconstruct the legal regulated rent on the base date may be applied to incorporate increases (whether fraudulent, erroneous or simply lacking in adequate documentation many years after the fact) in the apartment's distant rental history, thereby expanding a tenant's total overcharge recovery well beyond what was provided under the prior law.

This retroactivity period cannot be characterized as brief; rather, the legislature appears to have intended that the retroactive period be bounded only by the length of the apartment's rental history. Such a vast period of retroactivity upends owners' expectations of repose relating to conduct that may have occurred many years prior to the recovery period. Having reasonably relied on pre-HSTPA statutory and regulatory provisions to destroy records (see former RSL § 26-516 [g]; Cintron, 15 NY3d at 354; Thornton, 5 NY3d at 181; Matter of Gilman v New York State Div. of Hous. & Community Renewal, 99 NY2d 144, 149 [2002])—records that are now needed under the HSTPA to establish the legality of prior rent increases and a lack of willfulness—owners may be held liable under the HSTPA for purported historical overcharges that were once supported by documentation. Turning to the treble damages [*18]provisions, where owners are unable to meet their burden to prove a negative—lack of past willfulness—the HSTPA makes treble damages mandatory for all six years of the new recovery{**35 NY3d at 380} period, rather than the two years preceding filing of the complaint. These provisions either increase the penalty or impose a new penalty for damages that previously were not trebled.

There can be no doubt here that the HSTPA part F amendments represent a clear rejection of prior rent stabilization enforcement policy and effectuate a significant readjustment of substantive rights relating to overcharge recovery, distinguishing this legislation in critical ways from that applied to past conduct in Romein, which clarified the Michigan Legislature's original intent that was expressed in a legislative resolution but disregarded by employers and courts. As explained further below, "judicial confusion" regarding how to calculate overcharges in Roberts cases (dissenting op at —) cannot alone transform the substantive amendments made in part F as to all overcharge cases into mere clarifying amendments like those in Romein. In the same way, the part F amendments are quite different from the 1997 RRRA amendment adding the lookback rule to the RSL's enforcement provisions, which this Court applied to a pending case in Partnership 92 (11 NY3d at 860; see n 22, supra). As we explained in Thornton, that lookback rule amendment merely clarified past legislative intent and reinforced existing statutory language which, since 1983, made clear that damages could not be calculated based on an overcharge that occurred more than four years prior to the filing of the claim (5 NY3d at 180).^[FN26] No Landgraf analysis was necessary with respect to application of the 1997 lookback rule{**35 NY3d at 381} to pending cases because, unlike the sea change created by the HSTPA part F amendments, it did not have a truly retroactive effect on liability. Moreover, although we applied the lookback rule amendment to past conduct in Partnership 92, we were more circumspect with regard to other amendments in the 1997 RRRA. In Gilman, we held that DHCR acted irrationally when it applied an amendment relaxing evidentiary requirements for admission of owner records to permit an owner to reopen the record, nearly a decade after the tenant commenced the proceeding and during the administrative appeal, expressing concern that "the rules were changed in midstream" (99 NY2d at 147, 149-152). This Court's precedent regarding the 1997 RRRA is more nuanced than the [*19]dissent acknowledges and is compatible with our analysis identifying the significant retroactive effects that would arise if part F is applied to pending cases.

Indeed, the effects of the HSTPA amendments expanding overcharge liability implicate the concerns that the Supreme Court expressed in Eastern Enterprises in striking down retroactive application of a statute that required former mine operators to fund the health benefits of retired miners who worked for the operator before it left the industry (524 US 498). Even though the statute reflected similar policy goals as the scheme upheld in Turner Elkhorn, it was invalidated (by the plurality on Takings Clause grounds, with a concurrence on due process principles) based on the extreme degree and arbitrary nature{**35 NY3d at 382} of the retroactive effect (524 US at 530-537, 547-550).^[FN27] It is clear from Eastern Enterprises that there are limits on retroactive imposition of liability even when it is related to a rational statutory goal.

Moreover, retroactivity concerns are further heightened where, as here, the new statutory provisions "affect[ ] contractual or property rights, matters in which predictability and stability are of prime importance" (Landgraf, 511 US at 271). While the lease agreements between the owners and tenants were necessarily subject to the requirements of the RSL, curtailing the parties' freedom of contract in significant degree, when the governing law (essentially incorporated in the annual leases) is altered retroactively years later, long after the expired contracts have been performed, the impact on contract rights is unusually significant. Such alteration—if applied retroactively—impairs real property rights by diminishing or possibly eliminating the constitutionally protected return on investment owners realized in the past related to the use of their properties (see generally I. L. F. Y. Co., 11 NY2d at 492). The HSTPA does much more than require a party to shoulder a new payment obligation going forward—and its destabilizing effect is especially severe.

That potential effect is demonstrated by the cases before us. In Regina Metro., for example, as noted by the Appellate Division dissent, application of the standard calculation methodology under the former rule resulted in overcharge damages of $10,271.40, while the reconstruction method erroneously utilized by DHCR—which appears consistent with the HSTPA's new approach—resulted in damages of $285,390.39 (Regina Metro., 164{**35 NY3d at 383}AD3d at 433 [Gische, J., dissenting]). In Reich, proper application of the pre-HSTPA statutes resulted in no overcharge, but a comparison of the market rent actually charged during the recovery period—over $18,000 per month—against a reconstructed stabilized rent under the HSTPA considering rental history dating back to the tenants' initial occupancy of the apartment in 2005 (or before) could result in an enormous retroactive increase in liability. The same profound impact on overcharge calculations would occur in Taylor and Raden, involving tenants that took occupancy well over a decade before they sued.

Unlike cases where retroactive application rationally furthered a legislative goal, such as closing a state-administered fund benefitting insurers that imposed unsustainable costs on employers (see American Economy, 30 NY3d 136) or preventing legislation from being undermined by those seeking to escape its impact before enactment (see R. A. Gray & Co., 467 US 717)—there is no indication here that the legislature considered the harsh and destabilizing effect on owners' settled expectations, much less had a rational justification for that result. While prospective application of part F to overcharges occurring after the effective date may serve legitimate and laudable policy goals, no explanation has been offered, much less a rational one, for retroactive application of the amendments to increase or create liability for rent overcharges that occurred years—even decades—in the past.

[*20]Part F contains no statement of legislative findings. Such a statement is contained elsewhere in the legislation, noting the continuing housing emergency; the need "to prevent speculative, unwarranted and abnormal increases in rents"; the acute shortage of housing accommodations caused by high demand and decreased supply; and the need, with respect to those being charged market rents, to avoid profiteering and other disruptive practices (L 2019, ch 36, § 1, part G, § 2). Prospective application of part F could be understood to address these concerns by deterring future overcharges, but retroactive application to cases pending in the appellate pipeline does not do so; the HSTPA cannot deter conduct that has already occurred (James Sq., 21 NY3d at 250). Likewise, to the degree that prospective application of certain provisions of the HSTPA is justified because the legislature has concluded that those provisions will act to preserve the stock of stabilized housing or{**35 NY3d at 384} moderate rents going forward, retroactive application of the amendments to increase the amount of an overcharge judgment (or create overcharge liability where none existed) does not return apartments to rent stabilization or ensure the propriety of rents collected in the future. Rather than serving any of the policy goals of rent stabilization (which it would not), retroactive application of the overcharge calculation amendments would merely punish owners more severely for past conduct they cannot change—an objective we have deemed illegitimate as a justification for retroactivity (see James Sq., 21 NY3d at 249-250; see also Turner Elkhorn, 428 US at 17-18 ["we would . . . hesitate to approve the retrospective imposition of liability on any theory of deterrence or blameworthiness" (citations omitted)]).

The dissent asserts that the overcharge calculation amendments were intended to ameliorate the overcharges arising from deregulations later revealed to be improper by Roberts and to address post-Roberts judicial confusion regarding how to calculate such overcharges and that, thus, retroactive application of such amendments to pending cases is supported by a rational basis (see dissenting op at —, —).^[FN28] This argument is unsupported by the text of the statute or its legislative history, which makes no reference to Roberts. The amendments impact far more than overcharges associated with the "14-year period of unlawful deregulation involving 50,000 [Roberts] apartments" referenced by the dissent (dissenting op at —). The overcharge calculation amendments apply to all overcharge claims—not merely those flowing from an improper deregulation, much less a Roberts deregulation. Thus, not only is there no basis to conclude that addressing Roberts was the legislature's intent, neither would these broadly applicable amendments constitute a rational response to Roberts.

Relatedly, although the new treble damages provisions function distinctly from the integrated overcharge calculation provisions, retroactive application of any part F amendments that would newly impose treble damages for past conduct is also{**35 NY3d at 385} impermissible.^[FN29] Treble damages are generally viewed as punitive (Vermont Agency of Natural [*21]Resources v United States ex rel. Stevens, 529 US 765 [2000]; State of N.Y. ex rel. Grupp v DHL Express [USA], Inc., 19 NY3d 278 [2012]; see also Senate Introducer's Mem in Support, Bill Jacket, L 2019, ch 36, 2019 McKinney's Session Law News of NY, No. 3 at A-99 [describing treble damages as "punitive"]). They function as such in the RSL, under which actual damages are also available and there are no limitations on the amount of the annual overcharge that may be trebled (see Landgraf, 511 US at 281 ["Retroactive imposition of punitive damages would raise a serious constitutional question"]; dissenting op at — [identifying constitutional concerns with retroactive imposition of treble damages]).

The legislature is entitled to impose new burdens and grant new rights in order to address societal issues and, in enacting the HSTPA, it sought to alleviate a pressing affordable housing shortage that it rationally deemed warranted action. But there is a critical distinction for purposes of a due process analysis between prospective and retroactive legislation. As the Supreme Court has observed, retroactive legislation that reaches "particularly far" into the past and that imposes liability {**35 NY3d at 386}of a high magnitude relative to impacted parties' conduct raises "substantial questions of fairness" (Eastern Enterprises, 524 US at 534). In the retroactivity context, a rational justification is one commensurate with the degree of disruption to settled, substantial rights and, in this instance, that standard has not been met. Thus, the overcharge calculation and treble damages provisions in part F may not be applied retroactively, and these appeals must be resolved under the law in effect at the time the overcharges occurred. The parties' remaining arguments lack merit, are rendered academic or are otherwise unreviewable.

In an attempt to delegitimize our analysis by association, our three dissenting colleagues raise the ghost of Lochner v New York (198 US 45 [1905]), an outdated and long-discredited Supreme Court precedent that has nothing to do with retroactivity (dissenting op at —, citing Lochner). In Lochner, under the guise of due process analysis, the Supreme Court struck down economic legislation it viewed as unwise from a public policy standpoint (see Ferguson v Skrupa, 372 US 726, 730 [1963]). We agree wholeheartedly with the dissent that legislative judgments are presumptively constitutional and are subject to a rational basis analysis in which the policy preferences of judges have no role. Although the dissent repeatedly suggests otherwise, in stark contrast to the holding in Lochner, in this case we are not invalidating or "striking down" the overcharge calculation provisions in the HSTPA. The only question presented and resolved here is whether those provisions—whose validity is not otherwise at issue in these appeals—may be applied retroactively. The dissent never seriously engages with this issue or the substantial body of precedent governing it.

In this regard, the rational basis test—although extremely deferential—must be meaningfully applied to ensure basic principles of fairness and substantial justice, lest we abdicate our responsibility to the citizens of this state. As Justice Holmes wrote when dissenting in Lochner—espousing a view that later prevailed in the Supreme Court—

"the word liberty in the Fourteenth Amendment is perverted when it is held to prevent the natural outcome of a dominant opinion, unless it can be said that a rational and fair [person] necessarily would admit that the statute proposed would{**35 NY3d at 387} infringe fundamental principles as they have been understood by the traditions of our people and our law" (Lochner, 198 US at 76 [Holmes, J., dissenting] [emphasis added]).

The modern rejection of Lochner has never been understood to require courts to abandon "fundamental principles" of fairness—not even when reviewing economic legislation. There are few principles as fundamental or, in the words of the Supreme Court, as "elementary" or "deeply rooted" as the notion that government may not irrationally impose or expand liability for past conduct (Landgraf, 511 US at 265).

Indeed, the legislation imprudently struck down in Lochner was not retroactive at all—it merely set a prospective cap on bakers' working hours (198 US at 52). If that legislation had directed that the cap be applied retroactively, requiring recoupment of wages bakers earned years if not decades in the past when working excess hours (or profits their employers earned as a result of the productivity associated with those hours), we would certainly look upon such retroactive application with skepticism. While it may be unusual—but not unprecedented (see Moe v Sex Offender Registry Bd., 467 Mass 598, 6 NE3d 530 [2014] [retroactive application of amendments [*22]requiring publication of sex offender registry information violated due process rights of sex offenders]; Neiman v American Natl. Prop. & Cas. Co., 236 Wis 2d 411, 613 NW2d 160 [2000] [retroactive application of amendment increasing cap on wrongful death damages violated defendant's due process rights]; San Carlos Apache Tribe v Superior Court ex rel. County of Maricopa, 193 Ariz 195, 972 P2d 179 [1999] [retroactive application of amendments revising surface water law violated due process rights of tribes])—to decline to apply a statute retroactively on due process grounds, it is also unusual for parties to ask the Court to apply retroactively legislation that alters substantive rights in the way that part F does.

We may have a fair disagreement over whether there is a rational justification for retroactive application of the HSTPA's overcharge calculation provisions, but the dissent's misguided attempt to cast as improper our application of a meaningful standard of constitutional review merits a response. We are persuaded by the words of Justice Breyer who, although disagreeing with the result in Eastern Enterprises, cautioned against the misplaced fear that reliance on the Due Process Clause in assessing the propriety of retroactive application of a{**35 NY3d at 388} statute somehow "resurrect[s] Lochner" (524 US at 557 [Breyer, J., dissenting]).

"[A]n unfair retroactive assessment of liability upsets settled expectations, and it thereby undermines a basic objective of law itself.

"To find that the Due Process Clause protects against this kind of fundamental unfairness—that it protects against an unfair allocation of public burdens through this kind of specially arbitrary retroactive means—is to read the Clause in light of a basic purpose: the fair application of law, which purpose hearkens back to the Magna Carta. It is not to resurrect long-discredited substantive notions of 'freedom of contract' " (id. at 558 [citations omitted], citing Ferguson, 372 US at 729-732).

Our "Court . . . plays a crucial and necessary function in our system of checks and balances. It is the responsibility of the judiciary to safeguard the rights afforded under our State [and Federal] Constitution[s]" (People v LaValle, 3 NY3d 88, 128 [2004]). Our narrow holding here—determining that newly-enacted overcharge calculation provisions may not be applied retroactively—constitutes nothing more than an appropriate exercise of this quintessentially judicial authority.

Accordingly, in Regina Metro., the Appellate Division order, insofar as appealed from, should be affirmed, with costs to petitioner Regina Metropolitan Co., LLC, and the certified question answered in the affirmative; in Raden, the Appellate Division order should be affirmed, with costs, and the certified question not answered as unnecessary; in Taylor, the Appellate Division order should be modified, without costs, in accordance with this opinion and as so modified affirmed, and the certified question answered in the negative; and in Reich, the Appellate Division order should be affirmed, with costs, and the certified question answered in the affirmative.

Wilson, J. (dissenting). For the first time in its history, our Court has struck down, as violative of substantive due process, a remedial statute duly enacted by the legislature: part F, § 7 of the Housing Stability and Tenant Protection Act of 2019 (HSTPA) (L 2019, ch 36, § 1, part F, § 7). According to the majority, when our legislature stepped in to remedy the unlawful deregulation of tens of thousands of rent-regulated dwellings, occurring because of the 13-year lapse between the New York{**35 NY3d at 389} State Division of Housing and Community Renewal's (DHCR) erroneous statement of the law and this Court's correction of it, the legislature violated the United States Constitution by deciding that landlords could retain some, but not all, of the unlawfully obtained rent overcharges.

The majority's justification is that "the prior statutory scheme conferred on owners clear repose" to retain unlawful rent overcharges "that occurred, in some cases, many years or even decades before the HSTPA was enacted" (majority op at —). But the prior rent control law offered no clear repose; rather, it produced differing judicial and administrative interpretations about how to calculate rent overcharge awards for past conduct. The legislature stepped in and resolved that question, as is its right. Moreover, even had the prior statute granted "clear repose," the legislature remains free to alter damages awards for unlawful rents obtained by unlawful past conduct. The very "claim revival" jurisprudence cited by the majority establishes the legislature's right to do so.

One hundred and sixteen years ago, in People v Lochner (177 NY 145, 157 [1904]), our Court understood that the legislature, not the courts, is charged with making laws to advance the public welfare and that courts must give a wide berth to such legislative judgments, so long as they do not trample constitutionally protected rights. The United States Supreme Court reversed us. Time has not been kind to Lochner v New York (198 US 45 [1905]). It is regarded as one of the Supreme Court's most misguided decisions. The majority's description of it as "long-discredited" (majority op at —) is charitable.

With today's decision, the disgraced era of Lochner makes its tragic return home. To find portions of the HSTPA unconstitutional on substantive due process grounds, the majority has disregarded jurisdictional rules and prudential concerns. It proceeds to mischaracterize the HSTPA's express application to "claims pending" as rendering it a retroactive "claim revival" statute. Answering a question not raised below, the Court decides it with a legal analysis not argued by the parties, applied to imagined factual circumstances on a nonexistent factual record.

In wielding substantive due process as a sword to strike down remedial economic legislation, the majority vitiates the political choices of New York's Legislature in passing the HSTPA. The amendments to the legislative scheme surrounding rent stabilization reflect the legislature's judgment, approved{**35 NY3d at 390} by the Governor, about the consequences for landlords who have violated New York law. In place of that judgment, the majority has substituted its own: the Court must "safeguard" the "substantive" "contractual or property rights" of New York's landlords. Indeed, late in its opinion, the majority identifies the substantive rights protected by its resurrection of Lochner: the "substantive rights relating to overcharge recovery" (majority op at —). That is, the rights of landlords to retain illegal overcharges wrongfully obtained from tenants.^[FN1]

Make no mistake: the legislature unequivocally instructed that part F of the HSTPA was to apply to "claims pending." The majority admits the legislature "intended [part F] to apply to overcharge claims where the calculation issue remained unresolved as of the June 2019 effective date" (majority op at —). As much as the majority protests it is "not invalidating or 'striking down' the overcharge calculation provisions in the HSTPA" (majority op at —), it is striking down, as violative of substantive due process, the legislature's clear command: "This act shall take effect immediately and shall apply to any claims pending or filed on and after such date" (L 2019, ch [*23]36, § 1, part F, § 7).^[FN2] This is Lochner redux: a grotesque usurpation of the legislature's role in determining economic regulation when no fundamental rights are at issue.

Because part F of the HSTPA contains economic regulations that reflect a legislative policy judgment and do not infringe on fundamental rights, it should be evaluated under the well-settled rational basis standard (West Coast Hotel Co. v Parrish, 300 US 379, 391 [1937]). The rational basis standard is not demanding (see People v Knox, 12 NY3d 60, 69 [2009]). Indeed, it is "the most relaxed and tolerant form of judicial scrutiny" (Dallas v Stanglin, 490 US 19, 26 [1989]). Simply, courts are barred from declaring economic legislation unconstitutional{**35 NY3d at 391} under the Due Process Clause of the Fourteenth Amendment if the regulation is conceivably rationally related to a legitimate government interest and is neither arbitrary nor discriminatory (see Nebbia v New York, 291 US 502, 537 [1934]). By finding a due process violation here, the majority ignores nearly a century of Supreme Court precedent in which the Court applied that rational basis test to state regulations (see Ferguson v Skrupa, 372 US 726, 729 [1963] [collecting cases]).

Since 1937, the Supreme Court has never struck down an economic regulatory statute, duly enacted by a legislature, on substantive due process grounds. Neither had we, until now. Because economic regulations, such as the rent control regulations before us, are not subject to any sort of heightened scrutiny and readily pass the rational basis test, I dissent.^[FN3]

In 1894, reform-oriented Republicans took control of every branch of New York government, after years of Democratic dominance backed by the notorious Tammany Hall organization (see Paul Kens, Lochner v. New York: Economic Regulation on Trial 38 [1998]). When the legislature reconvened for its 1895 session, bakers on the Lower East Side were on strike over working hours and conditions (id. at 49). At the time, most bakeshops were housed in unfinished, stooped-ceilinged tenement basements. Workweeks were typically more than 70 hours, and in some cases over 100 hours, for less than $12 per week (before boarding costs that workers were required to pay) (id. at 13).^[FN4] The work was hot, grueling, unsanitary and unsafe. Although New York had enacted a statutory eight-hour workday in 1867, the law contained no enforcement mechanism and included a section providing that "no person shall be {**35 NY3d at 392}prohibited from working as many hours extra work as he or she may see fit" (id. at 26). The bakeshop law (codified, as relevant, at Labor Law § 110, as added by L 1897, ch 415) set a 10-hour per day and 60-hour per week limit for bakery employees. It also made any violation of the law a misdemeanor punishable by a $20 to $100 fine on first offense (People v Lochner, 73 App Div 120, 123 [4th Dept 1902]). The bakeshop law passed unanimously in the Assembly and Senate and was signed by the Governor that May.

A Utica bakeshop proved to be the law's downfall. Joseph Lochner, a longtime adversary of Utica's journeyman bakers' union, was arrested in April 1901 for violating the bakeshop law by allowing (or compelling) his employee, Aman Schmitter, to work more than 60 hours per week. It was Lochner's second violation of the law, for which he faced a fine of $50. After his conviction in Oneida County Court, Lochner argued on appeal that the bakeshop law prohibited him from freely entering into contracts, in violation of the Privileges and Immunities and Equal Protection Clauses of the Fourteenth Amendment, and the Due Process Clause of the State Constitution (Lochner, 73 App Div at 121). The Appellate Division sided with the State, holding that the statute was a valid exercise of the legislature's police power to create economic regulations and that the judiciary must not disturb such a regulation if it "really relates to and is convenient and appropriate to promote the public health" (id. at 124, quoting Matter of Jacobs, 98 NY 98, 110 [1885]).

We agreed. As we held, many states had adopted statutes to address working conditions in various industries, and the Supreme Court had regularly upheld them as not violative of the Fourteenth Amendment (Lochner, 177 NY at 148-149, citing Barbier v Connolly, 113 US 27 [1885] [upholding a San Francisco ordinance banning overnight work in public laundries]; Holden v Hardy, 169 US 366 [1898] [upholding Utah's eight-hour workday for mine workers]). The standard for Fourteenth Amendment review of a statute was plain: "If the act and the Constitution can be so construed as to enable both to stand, and each can be given a proper and legitimate office to perform, it is the duty of the court to adopt such construction" and "[i]t is not necessary to the validity of a penal statute that the legislature should declare on the face of the statute the policy or purpose for which it was enacted" (Lochner, 177 NY at 159 [citations omitted]). The Court needed only to find a{**35 NY3d at 393} conceivable way in which the statute was addressed to the benefit of the public, which the bakeshop law was.

The Supreme Court held otherwise, spawning the dominant canon of review for state economic regulation from 1905 until its demise in 1937, and the dominant anticanon for the 83 years since (see Jamal Greene, The Anticanon, 125 Harv L Rev 379, 418 [2011]). In Lochner v New York, the Court held that New York deprived bakers of the "the general right of an individual to be free in his person and in his power to contract in relation to his own labor" (198 US 45, 58 [1905]). The Court articulated a new, heightened protection for some class of economic liberty, including the right to contract freely:

"Statutes of the nature of that under review, limiting the hours in which grown and intelligent men may labor to earn their living, are mere meddlesome interferences with the rights of the individual, and they are not saved from condemnation by the claim that they are passed in the exercise of the police power and upon the subject of the health of the individual whose rights are interfered with, unless there be some fair ground, reasonable in and of itself, to say that there is material danger to the public health or to the health of the employ‚s, if the hours of labor are not curtailed. If this be not clearly the case the individuals, whose rights are thus made the subject of legislative interference, are under the protection of the Federal Constitution regarding their liberty of contract as well as of person; and the legislature of the State has no power to limit their right as proposed in this statute" (Lochner, 198 US at 61).

Justice Oliver Wendell Holmes, dissenting in an enviable 650 words, articulated the once and future position of the Court: "state constitutions and state laws may regulate life in many ways which we as legislators might think as injudicious or if you like as tyrannical as this, and which equally with this interfere with the liberty to contract" (id. at 75-76 [Holmes, J., dissenting]). Justice Holmes vigorously maintained that it is up to the people and their political representatives to determine the extent of economic regulation, not a constitutional question for courts.

In the roughly 30-year Lochner era that followed, an estimated 200 state statutes were found to be unconstitutional{**35 NY3d at 394} as violative of due process because they interfered with the right to contract—what would come to be known as the first wave of the "substantive due process" doctrine (see Erwin Chemerinsky, Constitutional Law: Principles and Policies § 8.2.2 [5th ed], citing Benjamin Wright, The Growth of American Constitutional Law 154 [1942]). In Block v Hirsh (256 US 135 [1921]), Justice Holmes, this time writing for the Court, drove a crack into Lochner's armor. During the pendency of a lease, the District of Columbia passed new legislation regulating rental property. The landlord objected to the application of the new legislation to the preexisting lease, claiming to do so would deprive him of due process. The Court upheld the statute, noting that "we have no concern of course with the question whether those means were the wisest, whether they may not cost more than they come to, or will effect the result desired" (id. at 158).

Unbowed by the Supreme Court's Lochner-era jurisprudence, and perhaps emboldened by Block, in 1933, this Court upheld a New York statute setting the price of milk against a due process challenge (People v Nebbia, 262 NY 259 [1933]). That proved the turning point. The next year, reviewing Nebbia, the Supreme Court retreated to its proper deferential posture:

[*24]"So far as the requirement of due process is concerned, and in the absence of other constitutional restriction, a state is free to adopt whatever economic policy may reasonably be deemed to promote public welfare, and to enforce that policy by legislation adapted to its purpose. The courts are without authority either to declare such policy, or, when it is declared by the legislature, to override it. If the laws passed are seen to have a reasonable relation to a proper legislative purpose, and are neither arbitrary nor discriminatory, the requirements of due process are satisfied, and judicial determination to that effect renders a court functus officio" (291 US at 537).

The coups de grâce came in West Coast Hotel Co. v Parrish (300 US 379 [1937]) and United States v Carolene Products Co. (304 US 144 [1938]), where the Court clarified that so long as legislation did not violate a specified constitutional right, "restrict[ ] those political processes which can ordinarily be expected to bring about repeal of undesirable legislation," or exhibit "prejudice against discrete and insular minorities," a {**35 NY3d at 395}reviewing court must defer to a legislature that employs reasonable means towards a legitimate purpose (Carolene Products, 304 US at 152-153 n 4; Nebbia v New York, 291 US 502, 537 [1934]).

The post-Lochner consensus has held for nearly a century, resting on two principles. The first principle is separation of powers. The legislature, not the courts, determines the extent of economic regulation aimed at goals like health, safety, prosperity and equity. The second principle is that the "freedom to contract" and associated economic liberties are not constitutionally protected rights. Legislation that threatens someone's pocketbook is not subject to any heightened constitutional scrutiny in the way that, for instance, legislation that discriminates based on sex or race is.

During World War II, civilian industries were mobilized for war. The construction workers who had not become soldiers were put to work making planes, munitions and other wartime necessities. Housing construction dramatically slowed (see Herbert Levy, Rent Control in New York City: Another Look, 47 NY St BJ 193, 194 [1975]). In the war's aftermath, New York City faced a severe housing shortage. Although the federal government had frozen New York City rents during the war, those controls were repealed in 1947, leaving military alumni and their booming families vulnerable (see New York State Division of Housing and Community Renewal, Office of Rent Administration, Rent Regulation after 50 Years: An Overview of New York State's Rent Regulated Housing 1993 [1994]). In 1949, the federal government empowered the States to enact rent control laws by giving States authority "to assume administrative control of rent regulation and the power to continue, eliminate or modify the Federal system" (id.).

Throughout the 1950s and 60s, New York City took charge of its rent-controlled housing, easing wartime-like rent control laws as it grew more prosperous and as more housing was built. However, by 1969, the City's housing crisis was once again dire: "the Vietnam War caused a steep rise in the rate of inflation and locally, housing production slumped. The overall vacancy rate which stood at 3.2% in 1965 fell drastically to 1.23% in 1968" (id.). War, once again, led to a rapid escalation in New York City rents, which encouraged the City to enact the Rent Stabilization Law of 1969 (RSL). That local law laid the{**35 NY3d at 396} groundwork for New York's rent stabilization scheme underlying the cases before us, first passed as the Emergency Tenant Protection Act of 1974 (L 1974, ch 576) and subsequently amended in 1983, 1993, 1997, and again in 2019.

The legislature's purpose in rent regulation is conceptually no different than in regulating the hours of bakers or, for that matter, in any law seeking to regulate the welfare of New Yorkers. At bottom, each of the many changes to the rent regulation laws has reflected a legislative judgment about how those benefits and burdens must be weighed so that New York does not slip back into the unregulated tenements of Lochner or the strictures of wartime rent control. The HSTPA is just the next set of changes that reflect a legislative response to the current state of New York's housing woes, akin to legislative acts in countless other fields.

As the majority acknowledges, "no party doing business in a regulated environment like the New York City rental market can expect the [Rent Stabilization Law] to remain static" (majority op at —). As with the workers' compensation system at issue in American Economy Ins. Co. v State of New York (30 NY3d 136 [2017]), "[t]he allocation of economic benefits and burdens has always been subject to adjustment," therefore rendering claimed rights [*25]to stasis "inchoate" (id. at 148 [citations omitted]). Neither landlord nor tenant has any fundamental right to the regulations of the moment, especially within a highly regulated industry such as rent stabilization (see Matter of Schutt v New York State Div. of Hous. & Community Renewal, 278 AD2d 58, 58 [1st Dept 2000] ["since rent regulation does not confer vested rights, petitioners' argument that the application of the RRRA's limitation period to pending cases violates due process by depriving them of the benefit of pre-RRRA rent regulation provisions law more favorable to their claims is without merit" (citation omitted)]). Rather, the legislature is free to calibrate its policy decisions to the needs of war, peace and everything in between, so long as its legislation is not irrational. That is the lesson of Lochner's interment.

The majority knows that "legislative judgments are presumptively constitutional and are subject to a rational basis analysis in which the policy preferences of judges have no role" (majority op at —). Under the majority's view, voiding sections of the HSTPA is not the product of heightened review but rather the product of the rational basis test "meaningfully applied" (id. at —), by which the majority means applied, for the first{**35 NY3d at 397} time since 1937, to strike down economic legislation making a policy choice about social welfare.^[FN5]

In order to justify "meaningful" application of the rational basis standard, the majority asserts that, unlike other economic regulation, the HSTPA threatens "substantive rights" (majority op at —, —, —, —, —, —) and "considerable reliance interests" (majority op at —). The majority insists it is not applying Lochnerian analysis, but then concludes that the HSTPA violates due process because its "impact on contract rights is unusually significant" (majority op at —). It is odd to refer to a landlord's retention of an illegal rent as a "contract right." Indeed, the "substantive right" to which the majority refers is not the right of landlords to earn a reasonable return, it is the right to keep rents collected in violation of the rent stabilization laws.^[FN6]

The majority tries to distinguish its holding from Lochner by asserting that the HSTPA is retroactive whereas the bakeshop laws were prospective (majority op at —). That completely misunderstands what makes Lochner odious. Lochner did not err because it found that wage and hour laws violated the freedom of contract when in reality they did not; it erred because it treated the freedom to contract as a right that could not be overcome by a legislature's rational attempt to make policy decisions that impaired the economic positions of some while benefitting others. Regardless, as explained at length in part IV infra and throughout, the majority's attempt to distinguish Lochner fails because retroactive legislation is subject to the same rational basis review as prospective legislation (see Landgraf v USI Film Products, 511 US 244 [1994]; American Economy Ins. Co. v State of New York, 30 NY3d 136 [2017]). Today's majority is analytically indistinguishable from Lochner: it applies a substantive due process analysis to invalidate a statute based on economic interests that the majority treats as if they were constitutionally protected rights, when they are not.{**35 NY3d at 398}

Using the instant cases to reanimate the dead hand of Lochner requires a couple of grisly maneuvers. First, we lack jurisdiction to address the HSTPA; second, for prudential reasons if nothing else, striking down a statute on substantive due process grounds when the argument is made for the first time in this Court without record support for the claimed burden and equitable factors (see e.g. majority op at — [citing HSTPA's effect on "considerable reliance interests"]) is both unwise and injudicious.

In these cases, we lack jurisdiction to consider the HSTPA. All four cases are in our Court on certified questions from the Appellate Division. In McMaster v Gould (240 NY 379 [1925]), we considered the very jurisdictional issue raised here: when a new statute is enacted after the Appellate Division sends a case to us via certification, do we consider whether the Appellate Division's decision was correct under the new statute or under the law as it was when the Appellate Division rendered its decision? Our decision was clear: "If the court below was [*26]right when it certified the question it is still right" regardless of any later changes to the statute (id. at 385). Lest there be any doubt, we very shortly before, in Robinson v Robins Dry Dock & Repair Co., explained that when an appeal comes to us in some way other than via a certified question (e.g., as of right from a double dissent or by a leave grant from a final judgment), "the appellate court may dispose of the case in accordance with the law as changed by the statute" (238 NY 271, 281 [1924]).

The majority's attempt to sweep away our long-standing precedents by asserting that the breadth of the Appellate Division's question determines whether we may apply a statute enacted after the Appellate Division's decision is utterly groundless (see majority op at — n 15). Appellate Division practice cannot overrule Court of Appeals precedent. McMaster directs that, when a certified question asks whether the Appellate Division order was properly made, we must answer that question: was the order proper at the time it was made?

The only case cited by the majority to assert that we have jurisdiction to reach the challenges to the HSTPA is Matter of Gleason (Michael Vee, Ltd.) (96 NY2d 117, 122 [2001]).{**35 NY3d at 399} However, Gleason was not before us on a certified question. It was a final decision as to which leave was granted, so it falls squarely under Robinson's rule, not McMaster's. The majority's further claim that certifying specific legal questions is a "largely abandoned practice" is irrelevant (majority op at — n 15; see Olsen v Town of Richfield, 81 NY2d 1024 [1993]; Flick v Stewart-Warner Corp., 76 NY2d 50 [1990]).

Even if we possessed jurisdiction to consider a constitutional challenge to the HSTPA, we should not invalidate a statute on substantive due process grounds when the argument is considered for the first time before us on an empty record.

A party seeking to invalidate a statute on substantive due process grounds bears the burden to prove that the legislature acted without a rational basis (see Usery v Turner Elkhorn Mining Co., 428 US 1, 15 [1976]). Because the HSTPA was enacted after the Appellate Division rendered its decisions in these cases, the parties' briefs in the lower courts, naturally, did not mention the statute and there was no evidence in the record concerning the statute—it did not exist. The majority emphasizes the burdens placed on landlords by the HSTPA without evidence that any substantial burdens exist in the real world. The Court now invalidates part F, § 7 of the HSTPA based on a hypothesized calamity, announcing the severity of the burden as a matter of law, substantiated by nothing. It appears that, although we lack the power to find facts, we have the power to imagine them.

The prudent course here would be to do as we did in Post v 120 E. End Ave. Corp.: remit these appeals to Supreme Court (or, in the case of Matter of Regina Metro. Co., LLC v New York State Div. of Hous. & Community Renewal, to DHCR) (62 NY2d 19, 29 [1984] ["The amended statute should be applied to this appeal but because the facts have not been developed, we reverse and remit the matter to the Supreme Court for further proceedings"]). In that way, the parties could develop a record that would allow the careful determination of standing, preservation and burden, issues that would be first determined in courts (or an agency) able to find facts and then could come to us on a record that frames our legal determination.

I turn, next, to the fundamental proposition underlying the majority's substantive analysis of the HSTPA: that the HSTPA{**35 NY3d at 400} is retroactive because it shifts the statute of limitations and revives previously extinguished claims. That is a false premise.

To understand why the HSTPA is not a retroactive "claim revival" statute, one must keep three time periods in mind: (1) the amount of time within which a tenant may challenge the unlawful deregulation of an apartment; (2) the amount of time for which a tenant can claim damages sustained as a result of an unlawful deregulation; and (3) the age of the records a court (or DHCR) can examine to determine what the rental rate would have been if an apartment had not been unlawfully deregulated.

As to the first time period, there is not, and there has never been, a time limit on when a tenant can claim that a unit has been unlawfully deregulated. Both before and after the HSTPA, tenants have always been able to challenge an unlawful deregulation of an apartment, no matter how far in the past the deregulation occurred (see e.g. Roberts v Tishman Speyer Props., L.P., 13 NY3d 270 [2009] [tenants brought suit in 2007 for an unlawful deregulation in 1993]; Kuzmich v 50 Murray St. Acquisition LLC, 34 NY3d 84 [2019] [tenants brought suit in 2016 for an unlawful deregulation in 2003]; Gersten v 56 7th Ave. LLC, 88 AD3d 189 [1st Dept 2011] [tenants brought suit in 2009 for an unlawful deregulation in 1999]).

As to the second time period, in 1983, the legislature set a four-year limit on rent overcharges that could be recovered as a result of an unlawful deregulation (see L 1983, ch 403, § 35). So, although a tenant could always seek a declaration that a unit was unlawfully deregulated 20 years ago, that tenant could recover overcharges sustained only in the four years prior to bringing the complaint. In 1995, if a tenant sued because a unit had been unlawfully deregulated in 1975, the tenant could claim damages for rent overcharges for the years 1991-1995. However, the court (or DHCR) could look back to 1975 to determine what the proper rent would have been for the years 1991-1995, had the unit not been unlawfully deregulated, and from that could determine the overcharge, if any, for those years.

As to the third time period, in the 1997 RRRA, the legislature amended the rent laws to limit courts (and DHCR) to looking back no more than four years from the filing of the complaint to determine the base rate from which the appropriate rental{**35 NY3d at 401} rate could be calculated (L 1997, ch 116, § 32; see Governor's Approval Mem, Bill Jacket L 1997, ch 116 at 40, 1997 McKinney's Session Laws of NY at 1923). Accordingly, as a result of the 1997 RRRA, a tenant who sued for overcharges in 2002 for an unlawful deregulation occurring in 1975 was limited to the use of records from 1998 or later to establish the rate that should have been charged from 1998 forward. Thus, even after the 1997 amendments, tenants could challenge an unlawful deregulation no matter how many years before that deregulation had occurred and could obtain both an injunction returning the apartment to rent regulated status and a measure of damages. Both before and after the 1997 RRRA, the period of allowable damages was four years, even if a landlord had unlawfully received decades worth of overcharges.

The HSTPA left the first time period unchanged. Tenants may still bring an action to declare that a unit was unlawfully deregulated at any time. The HSTPA lengthened the second time period, allowing the recovery of six years of overcharges instead of four. Even if a landlord has been overcharging tenants for decades, a tenant can still recover only a portion of the overcharge, though a larger fraction than before. Finally, the HSTPA eliminated the third time period altogether, by repealing the four-year lookback period embodied in the 1997 RRRA. That repeal allows the courts and DHCR to consider whatever evidence would best establish the rent had the unlawful deregulation never occurred. Simply put: pre-HSTPA, the remedy for an unlawful deregulation, from any time in the past, was four years of damages calculated in one way; post-HSTPA, the remedy is six years of damages calculated in a different way—the way they were calculated until 1997.

The majority's mischaracterization of the changes to the second and third time periods drives its retroactivity claim. Neither change created or extended a statute of limitations. From 1983 until 1997, if a tenant had suffered 20 years of illegal overcharges, the tenant could recover four years of damages determined (in some cases) by looking back 20 years to establish a base rate, carrying that rate forward, and applying it to the four-year period immediately preceding the complaint. From 1997 until 2019, tenants could still recover four years of damages, but the damages were cabined by requiring that the base rate could not be constructed by use of information more than four years before the complaint was filed. The cause of action stemming from the unlawful deregulation remained untouched.{**35 NY3d at 402}^[FN7] Now, through the HSTPA, the legislature has again tinkered with the recoverable amount and relaxed the evidentiary restriction somewhat. Doing so does not revive claims, nor does it make the HSTPA retroactive.

At no point during rent stabilization's long history could a landlord who unlawfully deregulated an apartment use the passage of time to escape an action for (1) a declaration that the apartment was unlawfully deregulated, (2) an order returning it to regulation, and (3) some measure of monetary damages. Therefore, there has never been a statute of limitations as to challenges to the wrongful deregulation of apartments. New York's constantly evolving rent laws [*28]have once again altered the remedy available to injured tenants, but the claim has always been the same and has never been subject to a limitations period.

The majority's analogy to claim revival statutes is inapt for the reason above—the HSTPA merely alters the fraction of overcharge damages recoverable by tenants and the evidence that may be used to prove the overcharge—and for another reason as well. Claim revival statutes, by definition, permit the assertion of claims that could not otherwise be brought. The claims here were all pending at the time the HSTPA took effect. The subject apartments were all unlawfully deregulated, for which the owners have been and remain liable. Indeed, by using "claims pending" to delineate cases that would be subject to the HSTPA, the legislature expressly did not revive claims that had been extinguished before the statute's effective date.

Comparison to the claim revival examples cited by the majority are instructive. The plaintiffs involved in the World Trade Center recovery efforts, whose claims were revived by Jimmy{**35 NY3d at 403} Nolan's law (L 2009, ch 440, reviewed in Matter of World Trade Ctr. Lower Manhattan Disaster Site Litig., 30 NY3d 377 [2017]), had been completely barred from filing any claims by the statute of limitations' expiration. The same is true with the plaintiffs whose DES-related claims were revived by the Toxic Tort Revival Act (L 1986, ch 682, § 4, reviewed in Hymowitz v Eli Lilly & Co., 73 NY2d 487 [1989]). In contrast, the claims here are pending, live claims. The legislature did not "revive" them; it altered the measure of damages and evidentiary rules relating thereto.

Even if part F of the HSTPA could fairly be characterized as a claim revival statute, its intentional application to "claims pending" would survive the majority's test (see majority op at — ["When the legislature has intended to revive time-barred claims, it has typically said so unambiguously, providing a limited window when stale claims may be pursued"]). First, as the majority recognizes, the legislature gave each of the HSTPA's 15 parts "its own effective date provision" and included "claims pending" only in part F's effective date (majority op at —). That differentiation is telling; the legislature meant what it said with particularity as to part F.^[FN8] Second, the claims "revived" have a similarly narrow window to the claims in Hymowitz and Matter of World Trade Ctr.: only two years of additional damages that would not have been recoverable under the old statute are available to these tenants. Damages covering the period from mid-2015 through June 14, 2019 and beyond were recoverable prior to the passage of the HSTPA and have not in any way been "revived" merely because the courts can look to older evidence to see if and by how much a tenant had been harmed during that period.{**35 NY3d at 404}

The Supreme Court does have a body of law directly on point, concerning the application of newly-enacted legislation to pending cases. The majority ignores that law, preferring instead to misconstrue the Court's decision in Landgraf v USI Film Products (511 US 244 [1994]), even though Landgraf pointedly reminds us that "the constitutional impediments to retroactive civil legislation are now modest" (id. at 272 [emphasis omitted]). Landgraf, read correctly and in the context of the larger body of the Supreme Court's decisions, demonstrates the unassailability of the legislature's choice to apply part F of the HSTPA to "claims pending."

Federal law has long recognized that new legislation governs pending cases. In Thorpe v Housing Authority of Durham (393 US 268 [1969]), the Supreme Court of North Carolina had held that a new HUD regulation granting tenants additional rights did not apply to pending eviction proceedings. The United States Supreme Court reversed, reaffirming that "[t]he general rule . . . is that an appellate court must apply the law in effect at the time it renders its [*29]decision" (id. at 281). The Court grounded its holding on United States v Schooner Peggy (1 Cranch [5 US] 103 [1801]), in which Chief Justice Marshall explained:

"[I]f subsequent to the judgment and before the decision of the appellate court, a law intervenes and positively changes the rule which governs, the law must be obeyed, or its obligation denied. If the law be constitutional, . . . I know of no court which can contest its obligation. . . . [T]he court must decide according to existing laws, and if it be necessary to set aside a judgment, rightful when rendered, but which cannot be affirmed but in violation of law, the judgment must be set aside" (id. at 110).

The Supreme Court has repeatedly reaffirmed that rule. "[T]he dominant principle is that nisi prius and appellate tribunals alike should conform their orders to the state law as of the time of the entry. Intervening and conflicting decisions will thus cause the reversal of judgments which were correct when entered" (Vandenbark v Owens-Illinois Glass Co., 311 US 538, 543 [1941] [collecting cases]).

In Bradley v School Bd. of Richmond (416 US 696, 710-711 [1974]), the Court reasserted that principle in a factual context directly relevant here. While Bradley's case was pending on{**35 NY3d at 405} appeal, Congress enacted legislation allowing for awards of attorneys' fees in school desegregation cases. Reiterating its long-standing rule, the Court held that as to claims pending when new legislation takes effect "a court is to apply the law in effect at the time it renders its decision, unless doing so would result in manifest injustice or there is statutory direction or legislative history to the contrary" (id. at 711). Although the statute in Bradley did not say it was to apply to pending cases, the Court held that Bradley could recover fees under the new statute. The Court concluded that taxing defendants with attorneys' fees where none had previously been allowed did not rise to a manifest injustice, and the absence of a legislative directive barring application to pending cases required application of the standard rule.

As in Bradley, the HSTPA has not imposed any new basis for liability: unlawful deregulation of rent regulated apartments was prohibited before and after the new legislation. The amount recoverable by the plaintiffs changed, but, as applied to pending cases, that did not and does not render the statutes unconstitutional. Furthermore, unlike the Congress in Bradley, our legislature was vocal, not mute. The changes made by part F of the HSTPA were intended to "take effect immediately and shall apply to any claims pending or filed on and after such date" (L 2019, ch 36, § 1, part F, § 7). That language "unequivocally convey[s]" that the HSTPA was explicitly formulated to apply to "any claims pending," including the claims before this Court (see majority op at — see also Kimmel v State of New York, 29 NY3d 386, 393 [2017] [" 'the word "any" means "all" or "every" and imports no limitation' " (citation omitted)]). The majority ultimately agrees (majority op at — ["the legislature evinced a sufficiently clear intent to apply part F to timely pending claims"]).

Landgraf reinforces the above long-standing rule. The Civil Rights Act of 1991 "create[d] a right to recover compensatory and punitive damages for certain violations of Title VII of the Civil Rights Act of 1964" and "allow[ed] monetary relief for some forms of workplace discrimination that would not previously have justified any relief under Title VII" (Landgraf, 511 US at 247, 254). By contrast, the HSTPA creates no new cause of action; the exact same conduct has always been proscribed by New York law. We stated clearly in Roberts that it was, and always had been, illegal under the statutory language of the RSL to decontrol luxury apartments while receiving J-51 tax{**35 NY3d at 406} benefits (see 13 NY3d at 285-286; see also Becker v Huss Co., 43 NY2d 527, 542 [1978] [applying a new contribution requirement in Workers' Compensation Law to conduct that happened prior to the statutory change because "the amendment neither created a new right nor impaired an existing one"]).

That difference—the creation of a new legal obligation as distinct from an alteration in the amount of damages available—was one of the two pillars on which the Supreme Court rested its decision in Landgraf that section 102 of the 1991 Civil Rights Act did not apply to pending cases. The majority admits the importance of that distinction when it notes that the legislation involved in American Economy did not violate due process because it "subjected the insurers to the possibility of such future costs but did not impose new legal liability" (majority op at — n 18). Just as "[t]he insurers were always legally liable for the closed cases" there (id.), the landlords have always been liable for unlawful deregulations and rent overcharges here. Just as "the fund merely provided [insurers] potential relief from the uncertain future coverage costs" in American Economy (id.), the four-year limitation on recoverable damages and the evidentiary lookback period merely provided potential relief to landlords from future rent overcharge cases. The same logic from American Economy applies here and should lead us to the same conclusion we reached there, unanimously.

The other pillar on which Landgraf rested was that the prior iteration of the bill, which was vetoed by the President and as to which the attempted override failed, "contained language expressly calling for application of . . . the section providing for damages in cases of intentional employment discrimination, to cases arising before its (expected) enactment" (511 US at 255). The subsequent version that passed omitted that language and, in context, the Court concluded that the absence of a legislative directive to apply section 102 to pending cases counseled against doing so. Here, again, the facts are just the opposite: our legislature has told us that part F applies to "claims pending."^[FN9] {**35 NY3d at 407}

Putting aside, for the moment, the preceding errors, I turn now to the majority's conclusion that part F of the HSTPA, as applied to pending claims, violates substantive due process.

Why do parties seeking to invalidate economic regulatory statutes have such a high burden? Why does a party seeking to invalidate such a statute have to show that the legislature could have had no rational basis for enacting it? The "most relaxed and tolerant" rational basis test applied to substantive due process challenges is rooted in separation of powers doctrine. The New York State Constitution places legislative policy judgments squarely within the province of the legislature (NY Const, art III, § 1).^[FN10] Policy judgments passed into law by the legislature are afforded a strong presumption of constitutionality to ensure that courts do not usurp the legislature's function (see People v Knox, 12 NY3d 60, 68-69 [2009]; Montgomery v Daniels, 38 NY2d 41, 54 [1975]). Even when a particular regulation may not be wise, "it is for the legislature, not the courts, to balance the advantages and disadvantages of the new requirement" (Williamson v Lee Optical of Okla., Inc., 348 US 483, 487 [1955]; see also Defiance Milk Prods. Co. v Du Mond, 309 NY 537, 541 [1956] ["Questions as to wisdom, need or appropriateness are for the Legislature. Courts strike down statutes only as a last resort and only when unconstitutionality is shown beyond a reasonable doubt" (citations omitted)]).

Our Court is an unelected body, nominated by the Governor, confirmed by the Senate, and not subject to a popular vote. The{**35 NY3d at 408} authority granted by the People of the State of New York to legislate for their benefit and in their name does not reside with the members of this Court. Instead, the legislature and Governor are responsible for making the final policy judgments that become law, and this Court is charged with exercising great restraint before invalidating an expression of popular will.

Separation of powers concerns—and the meager rational basis test used to evaluate substantive due process challenges as a result of those concerns—apply with equal force to regulations that have a retroactive effect. A new provision may have retroactive effect if it "attaches new legal consequences to events completed before its enactment" (Landgraf, 511 US at 270-271). When a statute with retroactive effect faces a substantive due process challenge, "the test of due process for retroactive legislation 'is met simply by showing that the retroactive application of the legislation is itself justified by a rational legislative purpose' " (American Economy Ins. Co. v State [*30]of New York, 30 NY3d 136, 158 [2017], quoting Pension Benefit Guaranty Corporation v R. A. Gray & Co., 467 US 717, 730 [1984]).^[FN11] Although there is a presumption against retroactivity that applies to all legislative provisions, the legislature can negate that presumption merely by "explicitly stat[ing] or clearly indicat[ing]" its intent that a new provision in the law apply retroactively (Gleason, 96 NY2d at 122). In Gleason, the legislature amended a procedural requirement for reviewing arbitration awards, stating that "all subsequent applications shall be made" in the new manner (id. at 120). No other language suggested that the amended statute would apply retroactively. Yet that modest language, its sparse legislative history, and its remedial nature were sufficient to overcome the presumption against retroactivity and convince this Court that the amendment should be applied retroactively (id. at 122-123). Thus, even if we were to consider part F of the HSTPA as{**35 NY3d at 409} retroactive, the legislature has unequivocally indicated its intent that part F apply to "claims pending."

General Motors Corp. v Romein (503 US 181 [1992]) is particularly instructive. In 1981, the Michigan Legislature allowed employers to decrease workers' compensation payments made to disabled employees who were receiving benefits under the new, more generous state fund, enacted in 1980. Before the statute's effective date, employers began to reduce the payments they were making to previously-injured employees. The legislature did not amend the legislation, but instead responded with a joint resolution declaring that the 1981 law was not intended to apply to workers injured prior to the effective date. Despite the legislature's resolution, the Michigan Supreme Court sided with General Motors, permitting employers to reduce their payments to employees injured before the statute's effective date (Chambers v General Motors Corp., 422 Mich 636, 375 NW2d 715 [1985]). In response to that decision, the legislature amended the workers' compensation law, requiring General Motors to pay an additional $25 million to workers whose payments it had reduced. The United States Supreme Court upheld the amendment, even though General Motors was forced to pay for claims that had expired under the old law, because "[t]he purpose of the 1987 statute was to correct the unexpected results of the Michigan Supreme Court's Chambers opinion. The retroactive repayment provision of the 1987 statute was a rational means of meeting this legitimate objective: It preserved the delicate legislative compromise that had been struck by the 1980 and 1981 laws" (Romein, 503 US at 191).

Romein requires that we find part F constitutional as applied to pending cases. In Roberts, we held that under the RSL as enacted by the legislature, any building receiving J-51 benefits was not subject to luxury decontrol. Thereafter, a form of judicial chaos, or "understandable confusion" (majority op at —), ensued (see Matter of Regina Metro. Co., LLC v New York State Div. of Hous. & Community Renewal, 164 AD3d 420, 427 [1st Dept 2018] [limiting records to a four-year lookback in the case of a post-Roberts illegal rent]; Taylor v 72A Realty Assoc., L.P., 151 AD3d 95, 106 [1st Dept 2017] [refusing to limit records to a four-year lookback in the case of a rent overcharge because that provision would "essentially allow the owner to collect rent that might be in excess of what it could have otherwise charged plaintiffs, based upon its own misapprehension{**35 NY3d at 410} of the law"] [endorsed by the Attorney General and the DHCR]; John Manning Irrevocable Trust v Biggart, 2019 NY Slip Op 31256[U], *3 [Sup Ct, NY County 2019, Lebovits, J.] [explaining that the Taylor court's method of determining the base rent had been "repudiated" by the First Department]; Matter of 160 E. 84th St. Assoc. LLC v New York State Div. of Hous. & Community Renewal, 160 AD3d 474 [1st Dept 2018] [affirming the DHCR's use of a sampling method to determine a default base rate]; 125 Ct. St., LLC v Sher, 58 Misc 3d 150[A], 2018 NY Slip Op 50092[U] [App Term, 2d Dept, 2d, 11th & 13th Jud Dists 2018] [freezing rent at the last registered rent and disallowing any increases]; Ferentinos v CF E 88 LLC, 2018 NY Slip Op 33247[U], *12-13 [Sup Ct, NY County 2018, Cohen, J.] [refusing to freeze the rent at the last registered rent because such a freeze would result in a windfall]).

Before the legislature stepped in by enacting the HSTPA, not only were trial courts and appellate panels within the First Department split, but both the Attorney General and DHCR came down on one side of the split—the side now rejected by the majority. Before the majority here resolved the split, the legislature did. It passed part F of [*31]the HSTPA and applied it to all pending claims, setting out clear rules for the courts in determining damages after the understandable confusion following Roberts (see Dugan v London Terrace Gardens, L.P., 177 AD3d 1, 9 [1st Dept 2019] [the HSTPA "resolve(s) a split in this Department as to what rent records can be reviewed to determine rents and overcharges in Roberts cases"]). Under Romein, clearing up judicial confusion about a prior statute meets rational basis scrutiny as to retroactive legislation (Romein, 503 US at 192).^[FN12]

{**35 NY3d at 411}A party challenging a statute as violative of due process bears the burden "to establish that the legislature has acted in an arbitrary and irrational way" (Turner Elkhorn, 428 US at 15). Here, not only does the majority strike down part F, § 7 of the HSTPA in the face of Romein's clear precedent, but it does so without requiring the parties challenging the HSTPA's constitutionality to prove anything.

The majority can find no solace in Eastern Enterprises v Apfel (524 US 498 [1998]). There, as the majority notes, the Supreme Court held that retroactive application of a coal miner health care benefit scheme was unconstitutional (majority op at —, —). But only Justice Kennedy, in a sole concurrence, believed that the law violated substantive due process. The plurality rested on a takings theory, pointedly noting that "this Court has expressed concerns about using the Due Process Clause to invalidate economic legislation" and quoting Williamson's postmortem for the Lochner era: "The day is gone when this Court uses the Due Process Clause . . . to strike down . . . laws, regulatory of business and industrial conditions" (Eastern Enterprises, 524 US at 537-538, quoting Williamson, 348 US at 488). The majority's attempt to use Justice Breyer's dissent in that case to support the use of a due process challenge to invalidate retroactive legislation is a farrago. In the passage that the majority quotes (majority op at —, quoting Eastern Enterprises, 524 US at 557 [Breyer, J., dissenting]), the four-Justice dissent was criticizing the plurality{**35 NY3d at 412} for trying "to torture the Takings Clause to fit this case" when the claim was properly analyzed under due process (Eastern Enterprises, 524 US at 556).^[FN13] Justice Breyer, speaking for himself and three other Justices, [*32]emphasized that Congress' decision to make an employer pay retroactively for past health costs of its employees did not violate the Due Process Clause. I agree with the majority that the HSTPA is constitutional unless it fails a due process analysis. I, along with eight members of the Eastern Enterprises Court, part with the majority in its application of a moribund 1920s version of due process analysis.

Furthermore, were one to accept the majority's assertion that "there may be some correlation between due process and takings analyses of retroactive legislation" (majority op at — n 27), settled takings jurisprudence would doom the majority's invalidation of part F, § 7. For a real property regulation to constitute a taking, it must entirely prevent the property from being economically viable (see Lucas v South Carolina Coastal Council, 505 US 1003, 1019 [1992] ["when the owner of real property has been called upon to sacrifice all economically beneficial uses in the name of the common good, that is, to leave his property economically idle, he has suffered a taking"]; Rent Stabilization Assn. of N.Y. City v Higgins, 83 NY2d 156, 173 [1993] ["Regulation of private property constitutes an unconstitutional taking if it denies an owner economically viable use of the property (a per se regulatory taking), or if it does not substantially advance legitimate State interests"]).

The majority next turns to our tax jurisprudence, which is equally unavailing as a basis on which to invalidate part F, § 7 of the HSTPA. In Matter of Replan Dev. v Department of Hous.{**35 NY3d at 413} Preserv. & Dev. of City of N.Y. (70 NY2d 451 [1987]), we adopted a "harsh and oppressive" standard when evaluating the retroactive application of a tax statute under due process. In Replan, a prior tax statute had exempted certain increases in the assessed value of property when the petitioner began renovating two buildings. We held that the petitioner "could not have justifiably relied" on the exemption and that the amendment in question did not unconstitutionally deprive the petitioner of due process (id. at 453). In doing so, this Court constructed a new test for tax statutes: "Retroactivity provisions in tax statutes, if for a short period, are generally valid, and ordinarily are upheld against due process challenges, unless in light of 'the nature of the tax and the circumstances in which it is laid', the retroactivity of the law is 'so harsh and oppressive as to transgress the constitutional limitation' " (id. at 455 [citations omitted]; see also James Sq. Assoc. LP v Mullen, 21 NY3d 233, 246 [2013]).

Replan's "harsh and oppressive" standard was lifted verbatim from the United States Supreme Court's holding in Welch v Henry (305 US 134, 147 [1938]). After we decided Replan, the Supreme Court overturned Welch (see United States v Carlton, 512 US 26, 30 [1994] ["The due process standard to be applied to tax statutes with retroactive effect . . . is the same as that generally applicable to retroactive economic legislation"]). Carlton made clear that cases like Welch, which applied a heightened rational basis scrutiny to tax statutes, "were decided during an era characterized by exacting review of economic legislation under an approach that has long since been discarded" (id. at 34 [internal quotation marks and citation omitted]). Thus, James Sq. mistakenly relied on Replan without noticing Carlton. We recognized as much in Caprio v New York State Dept. of Taxation & Fin. (25 NY3d 744, 752 [2015] ["the 'harsh and oppressive formulation . . . does not differ from the prohibition against arbitrary and irrational legislation that applies generally to enactments in the sphere of economic policy' "]). To the extent that Replan and James Sq. apply any form of heightened scrutiny to due process review of economic legislation, they are not good law.

The Supreme Court proclaimed that using "the vague contours of the Due Process Clause to nullify laws which a majority of the Court believe[s] to be economically unwise" is an {**35 NY3d at 414}abandoned practice (Ferguson, 372 US at 731 [internal quotation marks omitted]). Even retroactive statutes will withstand a due process challenge " '[p]rovided that the retroactive application of a statute is supported by a legitimate legislative purpose furthered by rational means' " (Carlton, 512 US at 30-31). Thus, contrary to the majority's bifurcation, there is no difference in the Supreme Court's [*33]jurisprudence when evaluating retroactive economic legislation and prospective economic legislation under due process: in both cases, courts are instructed to apply the rational basis test.^[FN14]

Putting aside, for a moment, that it would be the landlords' burden to demonstrate that the application of part F to "claims pending" could have no rational purpose, and also putting aside the complete absence of a record showing any irrationality in the legislature's action, the proper way to analyze part F under the Due Process Clause would be to examine each challenged element separately, asking whether it has been proved to serve no rational purpose (see Matter of Beary v City of Rye, 44 NY2d 398, 411 [1978] [deciding whether each provision of a statute had retroactive effect]; see also L 2019, ch 36, § 1, part M, § 28 ["If any provision of this act, or any application of any provision of this act, is held to be invalid, that shall not affect the validity or effectiveness of any other provision of this act, or of any other application of any provision of this act, which can be given effect without that provision or application; and to that end, the provisions and applications of this act are severable"]). In so doing, we must keep in mind that "[t]here is, of course, not only an exceedingly strong presumption of constitutionality . . . but a further presumption that the Legislature has investigated for and found facts necessary to support the{**35 NY3d at 415} legislation" (I. L. F. Y. Co. v Temporary State Hous. Rent Commn., 10 NY2d 263, 269 [1961]).

Court Reliance on Records More than Four Years Old

The HSTPA allows a court, in determining the lawful rent, to consider records older than four years before the last registration or annual report was filed; whereas the prior law (the 1997 RRRA) had limited courts to records within four years of a filed complaint. A plain rational purpose is evident: the legislature wishes courts to have access to all the evidence that is available to them when trying to establish what the regulated rent for an unlawfully deregulated apartment should be. That had been the law prior to the passage of the 1997 RRRA; the legislature merely changed an evidentiary rule about what courts may consider. By allowing courts to consider (or reject) old but reliable evidence of an overcharge, the legislature provided the return to a more flexible rule of evidence, and in so doing resolved a split within the First Department, deciding to side with the Attorney General and DHCR. That choice reflects a legitimate legislative purpose.

Change in Period Used to Measure Rent Overcharge Damages

The legislature chose to change the measure of damages for unlawful deregulation from a maximum of four years to a maximum of six years. That change represents its policy judgment about how much harm tenants living in illegally deregulated apartments have suffered (and should be compensated for) and how much landlords have improperly benefitted from unlawful deregulation while receiving J-51 tax benefits. It is important to remember that, on January 16, 1996, DHCR erroneously advised landlords that they were entitled to luxury decontrol of their apartments even while receiving J-51 tax abatements. Almost 14 years thereafter in Roberts, we held that DHCR was wrong, and all those apartments had been unlawfully removed from the rent-controlled housing stock within New York City (Roberts, 13 NY3d at 287). According to the majority, citing an amicus, 50,000 or more apartments were unlawfully removed from rent regulation between 1996 and 2009 because of DHCR's erroneous statement of the law (majority op at —). Surely, the legislature was free to reexamine the sufficiency of the four-year damages measurement period and decide, while not permitting full recovery back to the date of the illegal deregulation, something more was required to adjust the balance after the DCHR-induced mass removal of rent-regulated apartments and the attendant overcharges stretching,{**35 NY3d at 416} in many cases, much longer than six years. It is fair to say that [*34]the legislature, when it set the four-year period for overcharge damages, could not have imagined a 14-year period of unlawful deregulation involving 50,000 apartments.^[FN15]

In essence, both the 1997 RRRA's four-year measuring period and the HSTPA's six-year measuring period are akin to a liquidated damages statute. They both serve to render damages awards, often in amounts smaller than the actual damages sustained as a result of overcharges. Suppose the legislature said that in the event of a rent overcharge, in an effort to promote judicial economy and to avoid evidentiary problems, the landlord must pay damages to the tenant at a rate of $10 per square foot per year. If the legislature then changed the statutory award to $15 per square foot, because a lack of clarity in the law had caused a large, unexpected and protracted surge in illegal deregulations, it would be fully within its rights to do so. Simply because the measure used by the legislature is stated in terms of years, rather than dollars or square feet, does not render it unconstitutional (or a "revival" of claims). Look at it this way: any landlord who illegally deregulated an apartment 20 years ago, and who will now be liable for only six years of overcharges, has received a windfall. Any landlord who illegally deregulated an apartment five years ago, and is liable for only five years of overcharges, is merely refunding the unlawful overcharge in full. A legislative determination that, in view of the Roberts snafu, an increase in the allowable overcharge damages was necessary as to pending cases is exactly the sort of policy judgment with which we are forbidden to interfere.^[FN16]

{**35 NY3d at 417}Furthermore, the legislature's choice to alter the way in which overcharge damages are measured aligned with the Attorney General's, DHCR's, and First Department's reading of the 1997 RRRA in Taylor. It thus provided a "legislative judgment about what the law in question should be" (Gleason, 96 NY2d at 122). The plain language of the HSTPA, the urgency with which it was passed, its remedial nature, and its clear policy choice are more than sufficient to rebut any "deeply rooted" presumption against retroactivity and demonstrate its lawfulness.

Record Maintenance and Production

What the majority calls the HSTPA's "document retention policy" is not really that. Before the HSTPA's effective date, the statute read "[a]ny owner who has duly registered a housing accommodation . . . shall not be required to maintain or produce any records relating to rentals of such accommodation for more than four years prior to the most recent registration or annual statement for such accommodation." That language did not require a landlord to retain or produce anything; it barred the courts and DHCR from ordering a landlord to maintain or produce records dating more than four years from the most recent registration statement for an apartment.

The HSTPA altered that provision to read that a landlord

[*35]"shall not be required to maintain or produce any records relating to rentals of such accommodation for more than six years prior to the most recent registration or annual statement for such accommodation. However, an owner's election not to maintain records shall not limit the authority of the division of housing and community renewal and the courts to examine the rental history and determine legal regulated rents pursuant to this section" (RSL [Administrative Code of City of NY] § 26-516 [g]).{**35 NY3d at 418}

The first modification prohibits a court or DHCR from ordering landlords to produce or maintain records that are more than six years older than the most recent registration statement, but, like the prior version, does not require landlords to retain any records absent court or agency action.

Moreover, under both the old law and the HSTPA, the "record retention" period ran from the most recent registration, not from the time of a tenant's complaint. Prior to the HSTPA, a landlord who deregulated a unit in 2000 and stopped filing registrations (see majority op at —) could not be required to maintain or produce records prior to 1996 but could still be required to produce records that were 20 years old (or more) at the time of the complaint. Post-HSTPA, the same court could require the landlord to produce or retain records back to 1994. If, shortly after Roberts, a tenant filed suit for a 2000 deregulation, would the incremental difference between the 14 years of records (1996-2010) and 16 years of records (1994-2010) be so burdensome as to void this provision of the HSTPA?^[FN17] Although the majority speculates that landlords may have destroyed records the moment the four-year lookback period expired (majority op at —), landlords have numerous other reasons to retain records, including IRS audits (seven years); proof of depreciation of various assets (up to 40 years for residential rental real property); corporate record retention policies and the like. The majority's mere speculation that landlords may have destroyed records in reliance on the previous provisions of the rent regulation laws is surely insufficient to establish the existence of a fundamental right in those prior provisions. In any event, no landlord here has claimed that it failed to preserve old records, much less that it did so on reliance on this statutory provision.

The second modification expressly allows the courts and DHCR to examine the rental history even if a landlord has not{**35 NY3d at 419} maintained records. A rational basis for that change is obvious: if damages may now be calculated on a six-year basis, DHCR and the courts should not be constrained by the prior language limiting their authority to order maintenance or production of documents in a given case to four years. Further, the second modification makes clear that a landlord's failure to keep documents does not divest DHCR or the courts from using other information—for example, rental history—to attempt to arrive at the appropriate regulated rent figure. That proposition is hardly novel: courts and agencies, faced with the unavailability of information formerly held by a party, should use the best information available to reach a decision; that legislative purpose is self-evident.

Treble Damages for Willful Infringement

The HSTPA, like preexisting law, required the assessment of treble damages for a willful unlawful deregulation of dwellings. Under the prior law, treble damages were calculated based on two years' worth of overcharges, even though four years of overcharges could be recovered. Under the HSTPA, treble damages are measured against the full overcharge amount, that is, up to six years of overcharges.

Treble damages, as applied to willful violations of law, may serve several functions. They may serve as a strong deterrent, thus enhancing compliance with the law. They may serve a claim-pursuing function, encouraging vigorous challenges by private parties, which has both a deterrent and compensatory effect. They may also serve a punitive purpose directed at noxious violations of law (see e.g. Brunswick Corp. v Pueblo Bowl-O-Mat, Inc., 429 US [*36]477, 486 n 10 [1977] [antitrust]; Universal Health Services, Inc. v United States ex rel. Escobar, 579 US —, —, 136 S Ct 1989, 1996 [2016] [False Claims Act]; State of N.Y. ex rel. Grupp v DHL Express [USA], Inc., 19 NY3d 278, 286 [2012] [New York's False Claims Act]).

None of these defendants willfully violated the law by deregulating apartments according to then-existing DHCR guidance (see majority op at — RSL § 26-516 [a] [2] ["A penalty of three times the overcharge shall be assessed upon all overcharges willfully collected by the owner starting six years before the complaint is filed"]). All of the relevant plaintiffs have sought treble damages, but because they are not entitled to them as a matter of law (which the majority correctly recognizes as the consequence of Borden v 400 E. 55th St. Assoc.,{**35 NY3d at 420} L.P., 24 NY3d 382, 398 [2014]; see majority op at —), deciding the constitutionality of that provision is not necessary to deny their claims. We should not gratuitously hold a statutory provision unconstitutional when there is a nonconstitutional basis on which to resolve the issue (see People v Finkelstein, 9 NY2d 342, 345 [1961]).

That aside, the legislature could have made a rational determination that, whereas the vast majority of the illegal deregulations it sought to address were unintentional, because DCHR provided an incorrect legal interpretation, some landlords may have illegally deregulated apartments for reasons having nothing to do with reliance on DCHR's analysis, and may have known at the time that they were engaged in illegal deregulation. Because the legislature, in passing the HSTPA, decided not to allow tenants to recover their full overcharges, the legislature could rationally have determined that the overall cost of the illegal deregulation to the system should be borne differentially between the many landlords who relied on DHCR and the few who knowingly chose to violate the law. The treble damages, in this circumstance, would work as an allocation based on fault rather than as a penalty; such an allocation would be rational.

As to claims based on an unlawful deregulation occurring before the HSTPA's effective date, the deterrent rationale is absent (because the unlawful deregulation occurred prior to the HSTPA's enhancement of treble damages), and the claim-pursuing rationale is weakened though not absent (because the increase in treble damages provides a greater incentive to pursue pending actions vigorously, even though the claims have already been filed). The punitive rationale remains. Those rationales are sufficient to surmount a due process challenge on this vacant record.

However, New York courts have recognized that treble damages may be "punitive in nature and obviously designed to severely punish owners who deliberately and systematically charge tenants unlawful rents, while deterring other owners of stabilized premises who might be similarly inclined" (Matter of H.O. Realty Corp. v State of N.Y. Div. of Hous. & Community Renewal, 46 AD3d 103, 108 [1st Dept 2007]; see also Senate Introducer's Mem in Support, Bill Jacket, L 2019, ch 36). Such damages "share key characteristics of criminal sanctions" (Landgraf, 511 US at 281 [dicta]). If legislatively applied to past conduct, the imposition of treble damages may "raise a {**35 NY3d at 421}serious . . . question" under the Ex Post Facto Clause of the Federal Constitution (id.).

"The Ex Post Facto Clause of the United States Constitution prohibits states from enacting laws that criminalize prior, then-innocent conduct; increase the punishments for past offenses; or eliminate defenses to charges for incidents that preceded the enactment" (Kellogg v Travis, 100 NY2d 407, 410 [2003]). It applies to statutes that "seek to impose a punishment" (id.). Although it is usually applied to criminal statutes, the clause "may also be applied in civil cases where the civil disabilities disguise criminal penalties" (Louis Vuitton S.A. v Spencer Handbags Corp., 765 F2d 966, 972 [2d Cir 1985]). If the damages in a particular case are "punitive" or "exemplary" and are retroactively imposed (Landgraf, 511 US at 281), a civil court can evaluate whether the damages violate the Ex Post Facto Clause.

Whether the retroactive application of treble damages rose to such a level as to be punitive rather than redistributive might pose a constitutional problem under the Ex Post Facto Clause. Such a determination, of course, would need to await the award of treble damages in a particular case.^[FN18] Therefore, we can say only that retroactive application of these damages may pose a constitutional issue that lower courts can evaluate if relevant to some case at bar.

Mandatory Attorneys' Fees

[*37]Finally, the HSTPA made attorneys' fees mandatory in any case in which a landlord was found to have violated the statute. However, a change from permissive to mandatory fees does not expand the scope of the fees that can be collected or retroactively change liability for a landlord's illegal conduct (see Landgraf, 511 US at 270). In Bradley, the Supreme Court approved precisely this legislative change, allowing a new statute awarding attorneys' fees to apply to pending litigation (see 416 US at 724). Mandating attorneys' fees for pending actions easily passes rational basis scrutiny by providing an incentive for plaintiffs' counsel to pursue their pending claims more vigorously. The same purpose that satisfies rational basis scrutiny for mandatory awards of attorneys' fees in new cases—encouraging suits to combat unlawful deregulation in the face{**35 NY3d at 422} of New York City's "grave emergency" (RSL § 26-501)—meets that standard in pending cases as well. Those cases are "pending," not over and, indeed, may be far from over, as is the case with three of the four cases before us, even under the majority's holding. Through awards of attorneys' fees, the legislature provides incentives to tenants—not merely to file claims, but also to pursue them vigorously. Mandatory attorneys' fees are a time-honored way to provide that incentive.

The greatest irony attendant to the majority's opinion appears when one compares today's decision with our Court's treatment of the RRRA, which, in 1997, altered the rent regulation laws in a way that substantially favored landlords.^[FN19]

The legislature has judged the rent regulations laws, from their very inception through the HSTPA's amendments, to be both remedial and urgent (see Gleason, 96 NY2d at 122). RSL § 26-501 states that "a serious public emergency continues to exist in the housing of a considerable number of persons within the city of New York" and that "unless residential rents and evictions continue to be regulated and controlled, disruptive practices and abnormal conditions will produce serious threats to the public health, safety and general welfare" (id.; see also RSL § 26-502). The HSTPA, intending to update and refine the legislature's response to that persistent crisis, was likewise urgent.

The majority acknowledges that "the 'claims pending' language is sufficiently clear to evince legislative intent to apply the amendments to at least some timely overcharge claims that were commenced prior to enactment" (majority op at —). Nevertheless, the majority refuses to follow the legislature's{**35 NY3d at 423} direction to apply part F to pending claims. In stark contrast, we deferred to the 1997 New York Legislature when it restricted, rather than expanded, the types of evidence and, hence, amount of damages recoverable in rent overcharge complaints. In 1997, as a result of the RRRA, tenants were barred from introducing the rental history of an apartment prior to four years from when they made their rent overcharge claim. In 1996, more documentation had been allowed; in 1997, the documentation a tenant could introduce at trial was explicitly limited, curbing a tenant's ability to bring a rent overcharge claim and reducing the amount recoverable.^[FN20]{**35 NY3d at 424}

[*38]The 1997 RRRA stated that amendments such as this one "shall apply to any action or proceeding pending in any court or any application, complaint or proceeding before an administrative agency on the effective date of this act" (L 1997, ch 116, § 46 [1]). Therefore, we deferred. We simply looked to the language of the statute and held that, for example, "[b]y its terms, the Rent Regulation Reform Act of 1997 (L 1997, ch 116) applies to any proceeding that was pending before the New York State Division of Housing and Community Renewal at the time of its enactment" (Matter of Partnership 92 LP v State of N.Y. Div. of Hous. & Community Renewal, 11 NY3d 859, 860 [2008]). By the terms of the HSTPA, the legislature specifically included the word "pending," making clear that it did not want the new provisions in part F to only apply to future claims. Nevertheless, the majority refuses to defer, even though the HSTPA's expanded evidentiary provision that the majority uses to proclaim that landlords' rights have been violated is the exact mirror of the provision changed by the 1997 RRRA.

We have also deferred to changes in the Real Property Law that affected landlords and tenants. In 1983, the New York Legislature amended section 226-b of the Real Property Law (L 1983, ch 403, § 37). That amendment prevented a tenant from assigning his or her lease to another party without the landlord's consent, unless the tenant's lease stated otherwise. The amended section stated that "[t]he provisions of this section . . . shall apply to all actions and proceedings pending on the effective date of this section" (Real Property Law § 226-b [7]), and our Court did not hesitate to apply the statute to leases signed prior to the effective date of the amendment. Even though the tenant's rights under the lease were impaired by applying the new law to pending leases and lawsuits, we applied the statute to pending cases, as the legislature commanded, because the express language of the statute conveyed the legislature's clear intention (see Blum v West End Assoc., 64 NY2d 939, 941 [1985]; Vance v Century Apts. Assoc., 61 NY2d 716, 718 [1984]; Bennett v Rockrose Dev. Corp., 106 AD2d 256, 257 [1984], affd 64 NY2d 1155, 1156 [1985]; Fox v 85th Estates Co., 100 AD2d 797, 797 [1984], affd 63 NY2d 1009, 1010 [1984]; Sitomer v Melohn Props. Mgt., 108 AD2d 706, 707 [1985], affd 65 NY2d 881, 883 [1985]).{**35 NY3d at 425}

Going back to 1961, when the New York Legislature passed a law forbidding retroactive increases in rent, lawful under prior statute, our Court deferred. We easily held that there was a rational basis for passing the statute and that landlords did not have, in any particular rent control rule passed by the legislature, "an interest so vested as [*39]to entitle [them] to keep the rule unchanged" (I. L. F. Y., 10 NY2d at 270). The legislature had been clear that the new rules applied to pending proceedings (see id. at 266), and we once again deferred to the legislature.

Thus, in every prior case altering the benefits and burdens of the economic relations between landlords and tenants, we have never—until now—held that either group had a substantive right protected by the Due Process Clause, even when the legislative changes applied to pending cases in which the parties had relied on a prior incarnation of the legislation.

The majority has applied a different framework to the HSTPA than to every other economic and regulatory statute and abandoned jurisdictional and procedural rules along the way. The proud Court that recognized the legislature's right to address the crisis caused by tenement-based labor is here unrecognizable, as we deny the legislature the right to determine how best to address New York City's housing crisis. I may not have arrived at the plan devised by the legislature in the HSTPA. But that is not my job. Our Frankensteinian role in resurrecting Lochner by assembling ill-fitting fragments of moribund doctrines frightens me, because it portends ill for the future.

In Matter of Regina Metro. Co., LLC v New York State Div. of Hous. & Community Renewal: Order insofar as appealed from affirmed, with costs to petitioner Regina Metropolitan Co., LLC, and certified question answered in the affirmative.

Opinion Per Curiam. Chief Judge DiFiore and Judges Stein, Garcia and Feinman concur. Judge Wilson dissents in an opinion in which Judges Rivera and Fahey concur.

In Raden v W7879, LLC: Order affirmed, with costs, and certified question not answered as unnecessary.

Opinion Per Curiam. Chief Judge DiFiore and Judges Stein, Garcia and Feinman concur. Judge Wilson dissents in an opinion in which Judges Rivera and Fahey concur.{**35 NY3d at 426}

In Taylor v 72A Realty Assoc., L.P.: Order modified, without costs, in accordance with the opinion herein and, as so modified, affirmed, and certified question answered in the negative.

Opinion Per Curiam. Chief Judge DiFiore and Judges Stein, Garcia and Feinman concur. Judge Wilson dissents in an opinion in which Judges Rivera and Fahey concur.

In Reich v Belnord Partners, LLC: Order affirmed, with costs, and certified question answered in the affirmative.

Opinion Per Curiam. Chief Judge DiFiore and Judges Stein, Garcia and Feinman concur. Judge Wilson dissents in an opinion in which Judges Rivera and Fahey concur.

"(i) the rent charged on the base date cannot be determined; or

"(ii) a full rental history from the base date is not provided; or

"(iii) the base date rent is the product of a fraudulent scheme to deregulate the apartment; or

"(iv) a rental practice proscribed under section 2525.3(b), (c) and (d) of this Title [which concern conditional rentals designed to deprive tenants of the protections of rent stabilization] has been committed" (RSC § 2522.6 [b] [2]).

In such a case, DHCR sets the legal regulated rent using the lowest number resulting from four formulas, which include the sampling method (id. § 2522.6 [b] [3]). These RSC provisions are inapplicable by their terms in an overcharge case, such as a Roberts case, where the base date rent is the result of a mere mistaken overcharge (not fraud) and the rent charged on the base date is known.

"there are arguments showing that the business of debt adjusting has social utility, but such arguments are properly addressed to the legislature, not to us. We refuse to sit as a 'superlegislature to weigh the wisdom of legislation,' and we emphatically refuse to go back to the time when courts used the Due Process Clause 'to strike down state laws, regulatory of business and industrial conditions, because they may be unwise, improvident, or out of harmony with a particular school of thought' " (372 US at 729-732 [footnotes omitted]).

"Doubtless the legislative construction of the earlier statute is without binding force in any judicial proceeding. We cannot say that the Legislature has here attempted to interpose its authority upon the courts in regard to the construction of the earlier statute. In any matter brought before the courts in which rights of any party may be dependent upon the proper construction of the earlier statute, the courts are still free to place their own construction upon it, which may be contrary to the construction placed upon it by the Legislature. The validity of the provisions of the later statute does not depend upon the declaration of the legislative construction of the earlier statute. That declaration may be disregarded by the courts" (City of New York v Village of Lawrence, 250 NY 429, 447-448 [1929]).

Moreover, after the 1983 amendments, courts regularly interpreted the 1983 rent regulation as permitting DHCR and the courts to look back far beyond four years of rental records to determine the base rent and corresponding overcharges (see e.g. Matter of Vinsue Corp. v State of N.Y. Div. of Hous. & Community Renewal, 169 AD2d 592 [1st Dept 1991]; Carol Turner, 308 Realty v Spear, 134 Misc 2d 733 [Civ Ct, NY County 1987]). Had the courts genuinely misinterpreted the 1983 legislature's true intent, one would expect the legislature to have acted much more rapidly than the 14 years it took for the legislature to "clarify" its true intent (see Mayer v City Rent Agency, 46 NY2d 139, 149 [1978] ["The city council's carefully calculated characterization of Local Law No. 76 as 'clarifying' its intent when Local Law No. 30 was enacted must give way to the substantive fact that under Local Law No. 30 as originally enacted authorized rent increases based on labor pass-along were allowed in addition to annual 71/2% permitted increases, whereas Local Law No. 76 purported to place labor cost pass-along increases under the general 71/2% ceiling"]).