MEMORANDUM OPINION
SLEET, District JudgeL INTRODUCTION
This dispute asks the court to determine the constitutional limits to Delaware’s enforcement of its unclaimed property laws. “Unclaimed” or “abandoned” property refers to any property held by a business, where1 the business is not the owner of the property, and there has been no contact with the owner for the “dormancy period.” 12 Del. C. § 1198. Usually, unclaimed property is money in the form of savings accounts, checking accounts, stocks,' un-cashed dividend or payroll checks, traveler’s checks, unredeemed money orders or gift certificates, life insurance policies, etc.
All fifty states, as well as the District of Columbia, have laws addressing how businesses must handle unclaimed -property. Under Delaware law, the business is called a “holder.” 12 Del C. at § 1198(7). At the end of each year, holders must report and transfer (or “escheat”) to the state any property that has been abandoned as of the end of the prior year. 12 Del. C. § 1199. Ownership of the funds does not transfer to the state. Instead, the state holds the abandoned property in trust in perpetuity until claimed by the owner or the owner’s successor in interest. (A742). Until the owner claims the property, a state can use the funds for the general welfare.
The defendants are employees of the state government authorized to enforce Delaware’s unclaimed property laws, 12 Del. C. §§ 1101, et seq. (the “UPL”). Plaintiff Temple-Inland Inc. (“plaintiff’) is a Delaware corporation that manufactures corrugated packaging. (D.I. 113 at 2). Its principal place of business is in Texas, and its operations are primarily in Texas, Indianapolis, and Indiana. (Id. at 3). In 2008, defendants audited plaintiff for deficiencies *532in reporting and escheating unclaimed property for the previous 22 year time period.
Plaintiff claims that defendants’ audit violates certain provisions of the U.S. Constitution. Presently before the court are the parties’ cross-motions for summary judgment on plaintiffs claims for violation of substantive due process, the takings clause, and the ex post facto clause. Defendants have also asked the court to exercise its discretion in favor of abstention. (D.I. 112, 114). For the reasons set forth below, plaintiffs motion for summary judgment is granted-in-part and denied-in-part, and the defendants’ motion for summary judgment is denied-in-part and granted-in-part.
II. PROCEDURAL HISTORY
On May 21, 2014, plaintiff filed a complaint alleging violation of federal common law and several provisions of the U.S Constitution. (D.I. 1). On July 22, 2014, plaintiff moved for summary judgment on two of its six claims (violation of federal common law and ex post facto clause). (D.I. 22). At the same time, defendants moved to dismiss the entire complaint for failure to state a claim. (D.I. 24). The court granted defendants’ motion to dismiss the federal common law claim, and denied defendants’ motion as to the rest of the claims.1 (D.I. 38, 39). The court also denied plaintiffs motion for summary judgment, finding that issues of fact remained unresolved. (Id.). On December 3, 2015, plaintiff filed an amended complaint dropping its claims based on the commerce clause and the full faith and credit clause. (D.I. 108). Thus, the only remaining claims in this case are the ones currently before the court on the parties’ cross-motions for summary judgment.
III. BACKGROUND
A. Delaware’s Dependence on Unclaimed Property Revenue
Three historical factors have made a significant contribution to why the parties find themselves in this dispute: (i) Delaware’s dependence on unclaimed property revenue; (ii) the U.S. Supreme Court’s priority rules for escheating unclaimed property; and (iii) Delaware’s historically lax enforcement of its unclaimed property laws. The court provides this additional background for the purpose of placing this dispute in a broader context.
The state government has admitted that providing consumer protection is not its only goal in administering unclaimed property. (A593). Unclaimed property is Delaware’s third largest revenue source, making it a “vital element” in the state’s operating budget. (DE00001111).2 The difference between the amount collected by the state and the amount returned to owners shows that a large percentage of unclaimed property revenue remains with the state. For example, in 2007, approximately $364.9 million in unclaimed property went to the state’s general fund. *533(A588). But, around that same time, no more than $20 million was returned to owners.3 (A723).
The state government has also admitted that “[t]o the extent that future budget needs and growth are dependent on [unclaimed property revenue], it, too, must grow.” (DE00001111). Delaware’s approach to growing its unclaimed property revenue is significantly influenced by the priority rules set forth by the U.S. Supreme Court. (A598). Since states can use unclaimed funds for the general welfare, several states may make competing claims for the same property. But the U.S. Supreme Court has held that the same unclaimed property “cannot constitutionally be escheated by more than one State.” Texas v. New Jersey, 379 U.S. 674, 678-79, 85 S.Ct. 626, 13 L.Ed.2d 596 (1965). To resolve which state had priority, the U.S. Supreme Court established a set of rules under Texas v. New Jersey, 379 U.S. 674, 85 S.Ct. 626, 13 L.Ed.2d 596 (1965) and its progeny, referred to as the “Texas cases.” The “primary rule” gives the first opportunity to escheat to the state of the owner’s last known address as shown by the holder’s books and records. Delaware v. New York, 507 U.S. 490, 499-500, 113 S.Ct. 1550, 123 L.Ed.2d 211 (1993). If the primary rule fails because the holders’ records disclose no address for an owner, the “secondary rule” awards the right to es-cheat to the state in which the owner is incorporated. Id.
As defendants are aware, Delaware is legal home to a large share of the nation’s corporations. (See A593 (recognizing Delaware’s “unique circumstances”)). As a result, Delaware “receives a large share of all owner/address unknown abandoned property” generated in the United States. (A587). In fact, an estimated 90% of the unclaimed property collected by Delaware is owner/address unknown property. (A595). Thus, it benefits Delaware if the holders’ records do not have an address for the owner. The state government has recognized that “improved record keeping systems could reduce the level of owner unknown property,” and, if that happened, “Delaware could see its revenues reduced.” (DE00001115).
Despite the state’s dependence on unclaimed property revenue, it has historically been lax in enforcing its unclaimed property laws. Delaware, like most states, has voluntary compliance percentages in the single digits. (A645). Historically, Delaware has never received more than 14,000 unclaimed property reports a year, yet over 680,000 entities are incorporated in the state. ■ (A589, A692). Delaware has long known that holders’ low level of compliance is not due to “willful noncompliance,” but a lack of awareness of reporting obligations.4 (A595 n. 3).
As a result, a “significant portion” of Delaware’s unclaimed property revenue comes from its enforcement activities. *534(A596, A646). But even this has not been particularly robust. Delaware lacked a formal audit program for unclaimed property until 2001 and, as of 2008, that program was still in its “infancy.” (A588, A642). Delaware has a small staff dedicated specifically to unclaimed property and depends primarily on contractors to conduct its audits. (A640-41, A646). Because Delaware has limited resources for its enforcement activities, it targets for auditing “large companies” where “there is a greater chance of escheatable property being collected.” (A655).
B. The Audit
On December 22, 2008, plaintiff received notice that the defendants would be .conducting an audit to determine its compliance with the UPL. In the notice, the state’s Audit Manager said, “I’m sure all records are being retained under standard retention policies,” which defendants later admitted is typically 7 years. (A129, A27S, A455). The notice also stated that unclaimed property is to be reported to Delaware in accordance with the priority rules set forth in the Texas cases. (A455). That same instruction is provided in every Delaware audit notice letter and every Delaware Escheat Handbook. (D.I. 113 at 4)
On behalf of defendants, a contract auditor named Kelmar Associates, LLC (“Kel-mar”) audited two bank accounts from which plaintiff issued checks for accounts payable and payroll. (A465). The audit covered a 22 year time period—January 1986 to December 2007. (A459). But plaintiff was only able to produce complete records back to 2003 for accounts payable and 2004 for payroll. (D.I. 113 at 4). Plaintiff had disposed of older records in compliance with its retention policy. (A260-63). Plaintiff also produced all unclaimed property reports it filed in Delaware from 1998 to 2008, a couple of reports filed before 1998, and two audit reports by Texas, covering 1985 to 2005. (D.I. 113 at 4). For the several years in which plaintiff did not have complete records, defendants estimated the amount of unclaimed property.
C. Legislation Regarding the Use of Estimation
In 2010, the General Assembly amended 12 Del. C. § 1155 to provide that defendants can use estimation in unclaimed property audits. That statute reads in pertinent part, with key language highlighted:
Where the records of the holder available for the periods subject to this chapter are insufficient to permit the preparation of a report, the State Escheator may require the holder to report and pay to the State the amount of abandoned or unclaimed property that should have been but was not reported that the State Escheator reasonably estimates to be due and owing on the basis of any available records of the holder or by any other reasonable methbd of estimation.
The only restriction that Section 1165 puts on the use of estimation is that it be reasonable. Otherwise, defendants have discretion with how to employ estimation.
In 2001, the General Assembly considered, but did not enact, H.B. 617, which would have authorized the . use of estimation and establish a. five-year, document retention requirement for unclaimed property records. With respect to estimation, Section 6(b) proposed that:
In conducting an examination under this chapter, extrapolation or any other commercially recognized method of statistical sampling may be used if employed in a neutral manner that may establish the existence or non-existence or [sic] property subject to escheat and may otherwise be employed to the extent agreed upon by the escheator and the holder.
(A624). Section 6(d) addressed record retention, stating that:
*535Any records required to be maintained by the person under this chapter may be destroyed or otherwise disposed of five (5) years after the initial report is made, unless any year so encompassed is then the subject of an audit or investigation.'
(A624). Although the General Assembly eventually adopted a statute- permitting the use of estimation, it never adopted a record retention statute.5 Delaware’s Department of Finance has also not promulgated any guidelines or requirements regarding record retention, relying instead on “standard retention policies.” (A129, A275, A455).
D. Defendants’ Estimation Methodology
To help explain defendants’ estimation methodology, examples are taken from the accounts payable portion of Kelmar’s audit report for plaintiff. (A461-62). Kelmar calculated plaintiffs net liability for-each year it did not have records (i.e., the “reach-back years”) by'multiplying plaintiffs “adjusted calendar sales” for that year by a “ratio estimator.” The reach-back years were 1986-2002 for accounts payable and 1986-2003 for payroll. Plaintiff was credited any amounts it could prove it previously reported and paid for that calendar year (referred to as “prior filings”). (D.I. 126 at 2).
[[Image here]]
[Editor’s Note: The preceding image contains the reference for footnote6]
The ratio estimator is a fraction created from the information available in the years plaintiff had records, i.e. the “base years.” Here, the base years were 2003-2007 for accounts payable and 2004-2009 for payroll. A fraction is a numerator divided by a denominator. The numerator of the ratio estimator is the estimated “base period liability.” The denominator is plaintiffs adjusted calendar sales during the base period.7 Kelmar reduced the denominator to account for the growing use of electronic payments utilizing Automated Clearing House (“ACH”). This adjustment was based upon defendants’ view that ACH payments were less likely to be unclaimed property. (D.I. 115 at 9). The reduced denominator resulted in a larger ratio estimator, and thus, a larger amount of unclaimed property in the reach-back years.
[[Image here]]
The base period liability is the sum of:
(i) all unclaimed property reported to any *536state, (ii) any funds returned to the rightful owner not in the ordinary course of business, and (iii) any amounts not remedi-ated.
[[Image here]]
A check was considered returned not in the ordinary course if it was reissued and cashed by the payee after the audit began. Although plaintiff had a policy and practice of sending out due diligence letters to payees that had not cashed a check, Kelmar presumed that the checks would not have been replaced but for the audit. (A487-88). The parties dispute what caused the checks to be reissued in these instances.
To determine the unremediated amount for the base period, Kelmar conducted a sampling of aged checks. First, Kelmar identified all checks within the base years that had aged at least 90 days. (D.I. 113 at 5). Then, Kelmar excluded from this population any check already reported to Delaware or another state as unclaimed property, because they were already counted in the base period liability under “prior filed items.” (D.I. 115-4 at 47). Kelmar also excluded any checks with a value less than $20, because the low amounts were considered inconsequential. (Id. at 48; A483).
[[Image here]]
The population of checks to be sampled were separated into stratums defined by check value. From each stratum, Kelmar randomly selected sample checks, except all of the checks were selected from the stratum with the highest value. The selected checks comprised the sample set, which was 217 accounts payable checks totaling $6,112,942.10 and 140 payroll checks totaling $603,137.36. (A483, A490). The checks in the sample set were then researched for “remediation.”
[[Image here]]
An aged check was “remediated” when plaintiff was able to provide sufficient documentation to show that the check was not unclaimed property, either because it had *537been returned to the payee in the ordinary course of business or was not owed to anyone. (D.I. 115 at 8; A483). Of the sample set, plaintiff was unable to remediate four checks from accounts payable totaling $4,508.73 and three checks from payroll totaling $805.90. (A483, A490). All of the unremediated checks had an address, and only one payroll check, in the amount of $147.30, was . addressed to Delaware. (A483, A490). Kelmar used all of the un-remediated checks in sample set to estimate the total unremediated amount for the entire population of aged checks,8 (D.I. 113 at 5). , .
Most notably, Kelmar’s estimation methodology relies heavily on property escheatable only to other., states to increase the amount of unclaimed property owed to Delaware. Specifically, Kelmar included in its base period liability both unclaimed property reported to other states and unreme-diated checks with a known address that was not Delaware. A larger base period liability leads to a larger ratio estimator, which leads to a larger amount of estimated unclaimed property in the reach back years owed to Delaware. To illustrate the impact, the court calculates that if the amounts prior filed to'other states were removed from Kelmar’s calculation of base period liability, the ratio estimator would be 0.00114124%, not 0.00365273%. The lower ratio estimator would result in a lower net liability. Even if, for purposes of symmetry, credits of prior filed amounts to other states were removed from the calculation of net liability, plaintiffs liability for accounts payable would still decrease from $1,176,767.77 to $330,252.89.9
E. Plaintiffs Protest
At the end of the audit, plaintiff was assessed a liability in the amount of $2,128,834.13, comprised of $1,176,767.77 for accounts payable and $952,066,36 for payroll. (D.I. 115 at 9; D.I. 113 at 6-7). This means that plaintiff owed Delaware for each reach-back year an average of $69,221.63 for accounts payable and $56,003.90 for payroll. By comparison, following is a chart showing how much less plaintiff escheated to Delaware for the base years. Neither defendants nor Kel-mar suggested that plaintiffs reports during the base years were inaccurate. Indeed, the- estimation used during the audit assumes those numbers are accurate.
*538[[Image here]]
Plaintiff protested the assessment through several levels of administrative appeal, as set forth in 12 Del, C, § 1156. (D.I, 115 at 9). Under the administrative appeals process, the holder may first file a written protest with the state’s Audit Manager, The Audit Manager is allowed to consider only “those property types, amounts and issues related to the examination that are set out in the written protest of the holder.” 12 Del. C. § 1166(c). A holder may appeal the decision of the Audit Manager to the Secretary of Finance, who must appoint an independent reviewer to consider the appeal of the Audit Manager’s findings and make a written report to the Secretary of Finance. 12 Del. C. § 1156(f)-(g). Either the holder or the Secretary of Finance may, within 30 days after the Secretary of Finance mailed the independent reviewer’s written determination to the holder, appeal the independent reviewer’s written determination to the Court of Chancery, 12 Del. C. § 1156(c). “The Court’s review shall be limited to whether the independent reviewer’s determination was supported by substantial evidence on the record.” 12 Del. C. § 1156(j).
Here, plaintiff protested to the Audit Manager several aspects of the audit, including the use of estimation, (A478-571). The Audit Manager responded by directing plaintiffs attention to Section 1155, and stating that defendants were “forced” to estimate plaintiffs unclaimed property liability, because plaintiff “failed to maintain records sufficient ‘to permit the preparation of a report.’ ” (D.I. 115-4 at 56 (quoting 12 Del. C. § 1155)).
' Ultimately, the Audit Manager reduced the assessment by $95,435.39 because Kel-mar had double counted two sample checks. (D.I. 113 at 9). The independent reviewer further decreased the assessment, because he considered Kelmar’s adjustment to plaintiffs calendar sales based on the use of ACH payments to be unreasonable. (A-776-79). Specifically, the independent reviewer increased the amount of calendar sales, which reduced the ratio estimator, and in turn, reduced the assessment by almost a million dollars. (Id.). On April 22, 2014, the Secretary of Finance adopted the opinion of the independent reviewer and declared that plaintiff owed the state $1,388,573.97 in unclaimed property. (A760). Plaintiff did not file an appeal in the Court of Chancery and instead brought suit in this court challenging the constitutionality of defendants’ actions.
IV. STANDARD OF REVIEW
■ “The court shall grant summary judgment if the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law,” Fed. R. Civ. P. 56(a). The moving party bears the burden of demonstrating the absence of a genuine issue of material fact. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 586, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986). A party asserting that a fact cannot be—or, alternatively, is—genuinely disputed must support the assertion either by citing to *539“particular parts of materials in. the record,” or by “showing that the materials cited do not establish the absence or presence of a genuine dispute, or that an adverse party cannot produce admissible evidence to support the fact.” Fed. R. Civ. P. 56(c).
If the moving party has carried its burden, the nonmovant must then come forward with “specific facts showing that there is a genuine issue for trial.” Fed. R. Civ. P. 56(e). A factual dispute is genuine where “the evidence is such that a reasonable jury could return a verdict for the nonmoving party.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). The standard does not change merely because there are cross-motions for summary judgment. Appelmans v. Phila., 826 F.2d 214, 216 (3d Cir.1987); Krupa v. New Castle Cty., 732 F.Supp. 497, 506 (D.Del.1990) (“The filing of cross-motions for summary judgment does not require the court to grant summary judgment for either party.”).
V. DISCUSSION
Plaintiff has asked the court to grant summary judgment in its favor on its claims that defendants’ actions violated substantive due process, the takings clause, and the ex post facto clause of the U.S. Constitution. Defendants have asked the court to instead grant summary judgment in their favor or abstain based on the doctrine set forth in Railroad Commission of Texas v. Pullman Co., 312 U.S. 496, 61 S.Ct. 643, 85 L.Ed. 971 (1941). (D.I. 115 at 34-35). It is surprising that defendants ask for abstention now after almost two years of active litigation. Nothing prevented defendants from raising this request during the previous motion for summary judgment or motion to dismiss. Regardless, before addressing the merits of plaintiffs claims, the court will explain why Pullman abstention is not warranted.
A. Abstention
Defendants have asked the court to abstain under Pullman until a state court has the opportunity to interpret Section 1155, which they claim is an ambiguous statute. (D.I. 115 at 34-35). Pullman abstention is an “extraordinary and narrow exception” to a court’s duty to adjudicate a controversy properly before it. Planned Parenthood of Cent. N.J. v. Farmer, 220 F.3d 127, 149 (3d Cir.2000) (quoting Colo. River Water Conservation Dist. v. United States, 424 U.S. 800, 813, 96 S.Ct. 1236, 47 L.Ed.2d 483 (1976)). Pullman abstention is limited to those exceptional circumstances where construction of an ambiguous state statute by a state court could avoid or modify the federal question. Zwickler v. Koota, 389 U.S. 241, 248—49, 88 S.Ct. 391, 19 L.Ed.2d 444 (1967); Farmer, 220 F.3d at 150. Pullman abstention does not apply “if the state law is clear on its face, or if its meaning has already been authoritatively decided by the state courts, or if the constitutional issue would not be avoided or changed no matter how the statute is construed.” C. Wright, Law of Federal Courts at 304 (1983). The fact that a statute has never been interpreted by a state court does not, by itself,1 compel abstention. Farmer, 220 F.3d' at 150; Harman v. Forssenius, 380 U.S. 528, 535, 85 S.Ct. 1177, 14 L.Ed.2d 50 (1965). Both parties gave the issue of Pullman abstention only a cursory treatment. Nevertheless, the court’s analysis of the doctrine as applied to this case leads it to conclude that Pullman abstention is not appropriate here.10
*540In understanding when Pullman abstention is appropriate, the court finds helpful Smolow v. Hafer, 353 F.Supp.2d 561 (E.D.Pa.2005). In that case, plaintiff owned 300 shares of common stock that were canceled and delivered to Pennsylvania’s state treasurer, under the mistaken belief it was unclaimed property. Id. at 565. The treasurer converted the stock to cash, which then earned interest. Id. Upon plaintiffs request, the treasurer returned the amount received from the stock sale but, pursuant to his interpretation of a Pennsylvania unclaimed property statute, refused to transfer the interest. Id. No state court had construed the applicable statute.11 Id. at 572. When the plaintiff sued in federal court, the federal court abstained under Pullman because the statute was unclear and susceptible to an interpretation by a state court that could narrow or eliminate the federal constitutional question. If a state court construed the statute to mean that interest followed the principal, then the treasurer violated the statute, and the federal constitutional issue would never be reached. Id. at 574.
Here, in contrast, the statute is not ambiguous. Section 1155 clearly provides that the State Escheator has the authority to require a holder to pay the amount the State Escheator “reasonably estimates” to be due and owing. The fact that the statute does not require the State Escheator to precisely determine the amount due, but instead permits him to “reasonably estimate” the sum, does not render it ambiguous. See Ky. W. Va. Gas Co. v. Pa. Pub. Util. Comm’n, 791 F.2d 1111,1112 & 1118 (3d Cir.1986) (finding no ambiguity under Pullman abstention where statute required commission to determine whether proposed rates were “just and reasonable”); All. of Auto. Mfrs., Inc. v. Currey, 984 F.Supp.2d 32, 51 (D.Conn.2013) (declining Pullman abstention where statute required manufacturers to give dealers “fair and reasonable” compensation because “terms like ‘fair’ and ‘reasonable’ may be ‘inherently flexible,’ but they ‘are not inherently ambiguous’ ”). Because Section 1155 is unambiguous, there is no need for the court to exercise Pullman abstention in this case.
B. Substantive Due Process
The Due Process Clause provides that no state shall “deprive any person of life, liberty, or property, without due process of law.” U.S. Const. *541amend. XIV,- § 1. The core concept of due process is protection against “arbitrary” government action. Sacramento v. Lewis, 533 U.S. 833, 845, 118 S.Ct. 1708, 140 L.Ed.2d 1043 (1998). The criteria for determining what is fatally arbitrary depends on whether the plaintiff is challenging a legislative enactment or executive action. Id. at 846, 118 S.Ct. 1708. A legislative enactment is subject to rational basis review, meaning the statute will withstand a substantive due process challenge if the state “identifies a' legitimate state interest that the legislature could rationally conclude was served by the statute.” Nicholas v. Pa. State Univ., 227 F.3d 133, 139 (3d Cir.2000). Here, however, plaintiff is challenging executive action, specifically, defendants’ audit and assessment of plaintiffs unclaimed property liability. Executive action violates substantive due process protections “only when it shocks the conscience,” the meaning of which varies depending on the factual context. Ecotone Farm LLC v. Ward, 639 Fed.Appx. 118, 125 (3d Cir.2016) (quoting United Artists Theatre Cir., Inc. v. Warrington, 316 F.3d 392 (3d Cir.2003)).
There appears to be no precedent in this or any other court addressing whether a state’s executive action with respect to. unclaimed property shocks the conscience. The parties cited none and the court did not find any. This is unsurprising because most “shock the conscience” cases involve claims of excessive force or physical brutality. Sergio v. Doe, 769 F.Supp. 164, 167 (E.D.Pa.1991). The Third Circuit has not, however, limited the “shocks the conscience” standard to substantive due process claims based on physical harm. See Bridges ex rel. D.B. v. Scranton Sch. Dist., 644 Fed.Appx. 172, 2016 WL 953003, at *5 (3d Cir. Mar. 14, 2016) (stating that nonphysical types of harassment are subject to the shocks the conscience analysis); see also, e.g., Ecotone Farm LLC, 639 Fed.Appx. at 125 (applying shocks the conscience standard to elaim that government officials interfered with plaintiffs use and enjoyment of land); Kimbleton v. White, 608 Fed.Appx. 117, 119 (3d Cir.2015) (applying shocks the conscience standard to claim that government improperly suspended plaintiffs brokerage license). But, due to the unique nature of unclaimed property, the cases not involving physical harm do hot provide any useful analogies to the unclaimed property context.
Despite the lack of clear precedent, the court finds several aspects of defendants’ actions troubling. As discussed in more detail below, defendants waited'22 years to conduct an audit, avoided the otherwise applicable 6 year statute of limitations under dubious circumstances, gave holders ho notice that they would need to retain unclaimed property records to defend against unmeritorious audits, applied Section 1155 for a prolonged retroactive period for no obvious purpose other than to raise revenue, and failed to follow the fundamental principle of estimation where the characteristics of the sample set are extrapolated across the whole, which also puts plaintiff at risk of multiple liability. The parties did not necessarily address all of these issues, and for the ones they did address, chose to argue whether each one was ah independent violation of substantive due process. Lacking clear precedent, the court finds no need to determine if any of these actions on their own shock the conscience. It is sufficient that, in combination, defendants’ executive actions shock the conscience.
1. Statute of Limitations
The statute of limitations is a critical issue. It explains why defendants are allowed to assess a liability against plaintiff for something that happened 22 years *542ago.12 Delaware’ statute of limitations for state action against holders is set forth in 12 Del. C. § 1158. Section 1158 provides, in relevant part, that:
(a) The State Escheator, as soon as is practicable after receipt of any report required by this chapter, shall examine it to determine if it is correct. If the . Escheator finds that the report is .not correct, the Escheator shall notify the holder in writing by certified or registered mail of the amount of any underre-ported abandoned or unclaimed property due and owing. Notice of the proposed deficiency in payment , shall be mailed to the holder within 3 years from the date the report was filed.No suit to enforce the payment of a deficiency in payment of abandoned or unclaimed property shall be brought under § 1156 of this title against a holder unless the notice of deficiency in payment is mailed to the holder within the 3-year period provided in this subsection. In the case of an omission of abandoned or unclaimed property from a report having a value in excess of 25% of the amount of abandoned or unclaimed property disclosed in a report, a notice of deficiency in payment may be mailed to the holder within 6 years from the date the report was filed.
(b) If no report is filed, or if a false or fraudulent report is. filed with the intent to evade the obligation to pay over abandoned property, a notice of deficiency in payment may be mailed to the holder at any time.
12 Del. C. § 1158(a)&(b). Accordingly, the State Escheator normally has three years from the date a report is filed to bring an action to enforce payment of a deficiency. Id. If the deficiency is greater than 25% of the amount of abandoned property disclosed in a report, the State Escheator has six years, instead of three, to bring an action. 12 Del. C. § 1158(a). Finally, the State Escheator can bring an action “at any time” if no report is filed, or if a false or fraudulent report is filed with the intent to evade the obligation to pay over abandoned property.1312 Del. C. § 1158(b).
The governing statute of limitations is not, as defendants have repeatedly, asserted, set forth in 12 Del. C. § 1202. (D.I. 115 at 1, 17). Section 1202 is labeled “Periods of limitations not a bar” and provides that:
The expiration of any period of time specified by statute or court order, during which an action or proceeding may be commenced or' enforced to obtain payment of a claim for money or recovery of property, shall not prevent the *543money or property from being deemed abandoned property nor affect any duty to file a report required by this subchap-ter or to pay or deliver abandoned property to the State Escheator.
12 Del. C. § 1202. Accordingly, even when an owner’s claim against a holder is barred by a statute of limitations applicable to the dispute between the owner and holder, the holder is not relieved of its obligation to escheat the abandoned property to the state.-
Section 1202 does not address the time period in which the state must bring an action against a holder to enforce a deficiency payment. To the contrary, Section 1202 is designed to prevent the holder from taking ownership of unclaimed property rather than escheating it to the state. Put simply, Section 1202’s purpose is to prevent private escheat, whereby the holder rather than the state would enjoy the benefit of unclaimed property that could not be recovered by the owner. See People ex rel. Callahan v. Marshall Field & Co., 83 Ill.App.3d 811, 38 Ill.Dec. 944, 404 N.E.2d 368, 374 (1980); Staples, Inc. v. Cook, 35 A.3d 421, 424 (Del.Ch.2012); Benson v. Simon Prop. Grp., Inc., 281 Ga. 744, 642 S.E.2d 687, 691 (2007); Unif. Unclaimed Prop. Act § 19 cmt. (1995),
Here, defendants seek a deficiency payment for each year from 1986 to 2002. All of those years are more than 6 years past the date a report was or should have been filed. Thus, defendants must show they fit within the “at any time” exception of Section 1158. Defendants do not claim that plaintiff filed false or fraudulent reports. Instead, defendants have assumed that plaintiff had unclaimed property, but filed no report.14 This, of. course, ignores two other equally plausible explanations for why the reports could not be found, each of which would bar defendants’ action.
First, it is possible that plaintiff had no unclaimed property, and per- the defendants’ guidance, did not file a report. Defendants have consistently advised holders that only financial institutions have to file a negative report. (A382; A442). A negative report means the holder has no unclaimed property. When no negative report is filed, the statute of limitations never begins to run against defendants. The defendants have not warned holders of these consequences. •
Second, it is possible that plaintiff did in fact file a report, but it cannot be proven because the copies were lost or destroyed. In compliance with its record retention policy, plaintiff destroyed most of its unclaimed property reports after 10 years. (A262). For reasons that are not explained, it appears that defendants have likewise not kept copies of all the unclaimed property reports filed with the state. (A938 ¶¶ 5-7). Nevertheless, defendants have put the burden on plaintiff, not the state, to show that the reports were filed, so only holders, not the state, suffer negative Consequences from the failure to kéep all reports. Plaintiff has not raised an argument as to who should have the burden of showing that no reports were filed. But if the burden is properly on plaintiff, defendants have also not warned holders of the consequences from the failure to keep all of their unclaimed property reports for an unlimited period of time.
Given the fact that plaintiff was able to reproduce three reports from before 2002, and a report from every year within its record retention policy, it seems likely that plaintiff filed reports that are now missing. (A462). Thus, it is troubling that defen*544dants’ ability to seek a deficiency that is more than six years old depends on accepting only one explanation for why reports cannot be found (i.e., that plaintiff had unclaimed property and did not file a report). It is also troubling that defendants’ ability to advance this irrebuttable presumption is benefited by several of defendants’ own actions, including telling holders they do not need to file negative reports, .not telling holders they need to retain all of their unclaimed property reports, putting the burden on plaintiff to show they in fact filed reports with the state, and not keeping copies of all unclaimed property reports filed with the state.
2. Notice
Plaintiff argues that it had no notice that estimation would be used in the audit. (See D.1.113 at 29-33). Although it is doubtful that plaintiff had notice that estimation would be used, this misses the point as to what type of notice would have actually benefitted holders and mitigated any constitutional concerns. An auditor’s estimation depends entirely on the information available or not available from the holders’ records. Any disputes about how the estimation was conducted would ultimately have to consider whether the estimation reasonably reflected what was or should have been in the holders’ records. Thus, whether plaintiff had notice that estimation, in a general sense, would be used is not as important as notice that plaintiff would be left with few means to defend against overreaching in an unclaimed property audit unless it retained all of its records forever.
Here,..defendants never gave any notice that holders would need to retain records related to unclaimed property, Through its own research, the court has determined that when defendants sent their audit notice to plaintiff in December 2008, Delaware was one of only four states that had not adopted a record retention statute for unclaimed property records. (See Appendix A attached hereto). Forty-three states had adopted a record retention statute and an additional three states had a statute1 of limitations for state actions against the holder that effectively operated like a record retention statute, because it had no open ended exception like Delaware. (Id.). Accordingly, forty-six states expressly set a clear and finite number of years for which holders had to retain unclaimed property records. Of those forty-six states, forty-two required that records be maintained somewhere between 5 and 10 years. (Id.). Of the four outliers, no state required more than 15 years and a few states required as little as 3 or 4 years. (Id.). For the two states that required 15 years, the statutes had been on the books since at least 1982. (Id.). In fact, a significant majority of the states enacted record retention statutes before the turn of this century.
In spite of Delaware’s decision to march to the beat of a different drummer, defendants express surprise that plaintiff took a “calculated risk” of failing to anticipate an audit covering 22 years. (D.I. 115 at 17-18). The court cannot help but wonder at Delaware’s sense of cadence. As defendants admit, and academics agree, corporations do not normally retain records for longer than 7 years 'unless there is an identifiable need. (See A129 (State Escheator agreeing that companies typically have document retention policies of seven years); A275 (Delaware’s Audit Manager stating that the standard retention policy for companies is seven years); D.I. 115-5 at 34 (treatise stating that “[m]ost institutions do not maintain all of their records intact for periods in excess of six or seven years”)).
Defendants claim that they were not constitutionally required to give notice that holders should retain records. (D.I. 115 at *54521). But courts have relied on the existence of a record retention statute to conclude that a state has not unconstitutionally shifted the burden of proof. Plaintiff argues that defendants have the burden of proving that the property is, in fact, unclaimed before they can seek a deficiency payment, and estimation does not meet that burden. (D.I. 119 at 15, 25). It is true that, “[u]nlike taxation (where a state holds direct legal authority to collect, retain, and spend tax revenue for purposes of funding governmental operations), a state’s authority to receive and retain unclaimed property is derived from the owner’s rights in that property.” Houghton, et al., Unclaimed Property, 74-3rd C.P.S. (BNA 2014) (A812). Consistent with the principle that states have no greater rights than the owner, the U.S. Supreme Court has stated that the first step in determining which state may escheat unclaimed property is determining the precise relationship between the holder and owner.15 Delaware, 507 U.S. at 499, 113 S.Ct. 1550. Thus, it seems appropriate that defendants have the burden of proving that the property is in fact unclaimed. The court does not, however, agree with plaintiff that estimation could never meet this burden. .
In Illinois Physicians Union v, Miller, the government sought to recoup from plaintiff-physician excess compensation the government paid plaintiff under Medicare due to improper billing practices, and used estimation to calculate the amount of overpayment. 675 F.2d 151, 153 (7th Cir.1982). The Seventh Circuit rejected plaintiffs argument, similar to the one made here, that the government improperly shifted the burden of proof regarding the existence of overpayments to plaintiff by using estimation. Id. at 154. The court explained that the government satisfied its burden by relying on a rebuttable, as opposed to irre-buttable, presumption that the estimated amounts were overpayments. Id. The presumption could be rebutted because, as the court noted, physicians were statutorily required to keep the records necessary to fully disclose the extent of the services for which Medicaid reimbursement was sought. Id.
Other courts approving the government’s use of sampling, extrapolation, or estimation have similarly required that the plaintiff have an opportunity to rebut the estimation. See, e.g., Ratanasen v. Cal. Dep’t of Health Servs., 11 F.3d 1467, 1471 (9th Cir.1993). That opportunity is most readily provided where parties have notice that they should retain the records that the estimation is designed to represent. It seems the Delaware General Assembly was aware of the interdependent relationship between estimation and record retention when it initially considered adopting an estimation statute in 2001. In that sanie bill, the Delaware General Assembly also considered adopting a statute that holders should retain records for 5 years. (A621-24).
Finally, defendants’ reliance on Riggs National Bank is misplaced, because it does not support their proposition that the state is “not constitutionally required to give notice of the method it will use to calculate a civil liability that is already on the books.” (D.I. 1Í5 at 21). In Riggs, the District of Columbia enacted a statute in March 1981 mandating that holders report and deliver abandoned property to the district. Riggs Nat’l Bank v. Dist. Of Columbia, 581 A.2d 1229, 1233 (D.C.App.1990). The statute applied retroactively to all property which would have been presumed abandoned if the law had been in effect as of January 1, 1980, i.e., over a year earlier. Id. at 1234. The district picked this date *546precisely because “it approximates a time when the public was put on notice that the government was contemplating recouping unclaimed property.” Id. at 1234 n. 9 (emphasis added). This is exactly the opposite of what defendants claim Riggs represents.
3. Retroactivity
Plaintiff argues that defendants’ retroactive application of Section 1155 for a period of twenty-two years violates substantive due process. (D.I. 126 at 8). Defendants argue that retroactivity is not an issue in this case because they are not relying on Section 1155 as the, source of their authority to estimate. (D.I. 127 at 5-6; Tr. 57). It is difficult to comprehend how defendants can argue that they are not relying on Section 1155 when the Audit Manager specifically told plaintiff that Section 1155 authorized defendants’ use of estimation in plaintiffs unclaimed property audit. (D.I. 115-4 at 56). Moreover, what would have been the point of enacting Section 1155 if not to give defendants’ actions the cover of legitimacy? It would certainly make defendants actions appear even more arbitrary if they did not have a statute authorizing an estimation that can reach back twenty-two years. Defendants cannot evade the issue of retroactivity by claiming that they did not rely on Section 1155, and the court does not believe it a productive exercise to delve into the defendants’ subjective intent to determine when they are, or are not, relying on statutes that authorize their actions.16
Since the court has concluded that retroactivity is, indeed, an issue in the case, it turns now to plaintiffs argument that defendants’ prolonged retroactive application of Section 1156 violated substantive due process. (D.I. 119 at 23). One complicating factor is that courts typically use rational basis review to determine whether retroactivity violates substantive due process, because the statute itself sets forth how far back it applies. Here, however, defendants—not Section 1155—-set the length of the retroactive period. Accordingly, the court must consider whether defendants’ actions “shock the conscience,” not whether the General Assembly could rationally conclude that enacting Section 1155 served a legitimate state interest. The court, nevertheless, relies upon cases addressing retroactive legislation to help inform its own decision.
Plaintiff suggests that the sheer length of the retroactive period is a violation of substantive due process. But the length of the retroactive period does not, by itself, establish a violation of substantive due process.17 See Davon, Inc. v. Shalala, 75 F.3d 1114, 1126 (7th Cir. 1996) (finding no support for the proposition that a statute necessarily violates due process if *547it is retroactive for an unprecedented length of time); In re Chateaugay Corp., 53 F.3d 478, 491 (2d Cir.1995) (“The proposition that the Due Process Clause imposes a time limit on a law’s retrospective effect elicits no support from the case law_”). Instead, it appears that the court must consider whether the length of the retroactive period is appropriate given the reason why it is being applied retroactively.
Here, defendants’ purported reasons for applying Section 1155 retroactively do not withstand scrutiny. Defendants are correct that retroactive legislation shifting the benefits and burdens of economic life between different public constituencies is “presumptively constitutional.” Usery v. Turner Elkhorn Mining Co., 428 U.S. 1, 15, 96 S.Ct. 2882, 49 L.Ed.2d 752 (1976). But, defendants’ audit of plaintiff did not shift the benefits and burdens of economic life between different public constituencies. A survey of cases upholding “burden shifting” legislation helps illustrate the point.
In each of these cases, the statutes imposed liability on companies for harm they caused in the past, and profited from, which affected innocent parties who might not otherwise have the financial means to rectify that harm. See, e.g., Turner Elkhorn, 428 U.S. at 6-8, 96 S.Ct. 2882 (requiring coal mine operators to pay health benefits to coal miners suffering from black lung disease); Pension Ben. Guar. Corp. v. R.A. Gray & Co., 467 U.S. 717, 731-32, 104 S.Ct. 2709, 81 L.Ed.2d 601 (1984) (ensuring that employers fully funded pensions so that employees were not deprived of anticipated retirement benefits); Commonwealth Edison Co. v. United States, 271 F.3d 1327, 1329-30 (Fed.Cir.2001) (raising funds from utilities to decontaminate and decommission government facilities used to enrich uranium that utilities used to sell energy to customers). Defendants cannot show a similar one-to-one shift between the burden on plaintiff and the benefit to owners of unclaimed property.18 Defendants may collect a greater amount of money through their application of Section 1155, but owners are not entitled to share in that increase. Instead, the amount of unclaimed property an owner can collect remains unchanged, because they are entitled to only the amount they abandoned. No more; no less,
Similarly, defendants’ actions do not reinforce the viability of the state’s unclaimed property program. An animating concern behind the Supreme Court cases upholding “burden shifting” legislation is that the administrative programs were experiencing a financial shortfall that could have jeopardized the viability of the program. See, e.g., Turner Elkhorn, 428 U.S. at 8, 96 S.Ct. 2882 (enacting legislation to address “the insufficiency of state compensation programs” for black lung disease); Pension Ben. Guar. Corp., 467 U.S. at 731-32, 104 S.Ct. 2709 (acting to address concern’that a significant number of pension plans were “experiencing extreme financial hardship”).' Here, owners are not at risk of being unable to recover their unclaimed property, because Delaware’s unclaimed property fund is not experiencing a financial shortfall. Indeed, Delaware has collected significantly more unclaimed property than it returns to owners. (A588, A723).
Finally, defendants’ use of estimation will not make it easier for owners to be *548reunited with their abandoned property, and may, in fact, make it harder. Unclaimed property recovered by estimation necessarily does not identify the owner, the property, or the property’s value. But the state’s instructions for recovering unclaimed property provide that “[i]f property is not found in [the owner’s] name,” then “a claim cannot be initiated.” (D.I. 115-5 at 16). There is evidence suggesting that the lack of names makes it harder for owners to claim abandoned property. (See A574-79 (defendants discussing difficulties of resolving owner’s claim because escheatment was not through an itemized list identifying individual property but a bulk settlement amount); A300-02 (defendants unable to give an example of when estimated property was returned to an owner)).
Defendants blithely assert that if an owner cannot claim abandoned property because it is estimated, it is still better that the estimated property is used to serve the public good than to increase the profit margins of Delaware corporations. (D.I. 115 at 13). It is true that unclaimed property should not become a windfall for holders. See, e.g,, In re Monks Club, Inc., 64 Wash.2d 845, 394 P.2d 804, 806 (1964). But, at the same time, unclaimed property laws were never intended to be a tax mechanism whereby states can raise revenue as needed for the general welfare. (A35). States, violate substantive due process if the sole purpose of enacting an unclaimed property law is to raise revenue. Memo Money Order Co. v. Sidamon-Eristoff, 754 F.Supp.2d 661, 678 (D.N.J.2010) (finding that unclaimed property laws enacted with the intent to raise revenue for the state’s general fund did not violate substantive due process “so’ long as revenue raising was not the only basis for the legislation” (emphasis in original)); N.J. Retail Merchants Ass’n v. Sidamon-Erristoff, 669 F.3d 374, 398 (3d Cir.2012) (“[E]ven if revenue-raising was the primary purpose behind enacting [an unclaimed property law], as long as it was not the only legitimate purpose underlying the legislation, [the law] will pass rational basis examination.” (emphasis added)).
Unfortunately, defendants offered no credible reason for using estimation as it did in plaintiffs audit other than to raise revenue. Defendants weakly assert that estimation permits an owner, otherwise prevented by a statute of limitations from recovering abandoned property directly from the plaintiff, to obtain that property if it is in the custody of the state. (D.1.115 at 13). But Section 1202 already serves this very purpose. Section 1202 provides that a statute of limitations between the holder and owner does not excuse the holder’s duty to escheat the abandoned property to the state. 12 Del. C. § 1202. As a result, an owner can recover indirectly from the state what it could not recover directly from plaintiff. Estimation has nothing whatsoever to do with this process.
4. Estimation
Courts have routinely upheld the government’s use of statistical sampling as a valid audit tool, provided it was properly performed. Chaves Cty. Home Health Serv., Inc. v. Sullivan, 931 F.2d 914, 919 (D.C.Cir.1991) (noting that statistical sampling is routinely permitted where case-by-case review would be too costly); Balko & Assocs., Inc. v. Sec’y of Health & Human Servs., 555 Fed.Appx. 188, 194 (3d Cir.2014) (using statistical extrapolation to determine amount of Medicare overpay-ments); United States v. Ukwu, 546 Fed.Appx. 305, 308, 310 (4th Cir.2013) (upholding the use of statistical sampling to prove amount of loss in tax fraud case).
An estimation is properly performed when it is based on the principle that the unclaimed property in the reach-back years has “all the same qualities and characteristics” as unclaimed property in the *549base-years. See United States v. Mehta, 594 F.3d 277, 283 (4th Cir.2010) (explaining that an extrapolation estimates the values of a function or series outside a range based on the assumption that the trends followed inside the range continue outside it); In re Chevron U.S.A., Inc., 109 F.3d 1016, 1019-20 (5th Cir.1997) (“The essence of the science of inferential statistics is that one may confidently draw inferences about the whole from a representative sample of the whole”). Accordingly, due process violations arise where the estimation methodology creates misleading results. See Daytona Beach Gen. Hosp., Inc. v. Weinberger, 435 F.Supp. 891, 900 (M.D.Fla.1977) (finding denial of due process where sample size was too small); Yorktown Med. Lab., Inc. v. Perales, 948 F.2d 84, 90 (2d Cir.1991) (finding no due process violation from use of extrapolation where there was a “low risk of error”).
Here, defendants acknowledged this essential principle, but failed to apply it. (A743). Defendants admit that “for the periods where records do not exist, we were estimating for all states. Not just Delaware. And we don’t just use Delaware property to estimate for periods where records do not exist. We use property from every state.” (A295). Defendants do not dispute that, under the primary rule of the Texas cases, Delaware cannot escheat any checks from the base years with payees in other states. But defendants estimated that plaintiff owed unclaimed property to Delaware by extrapolating those checks into the reach back years.
Defendants’ actions are premised on the logic that they do not need to extrapolate the characteristics of the property on which the estimation is based, because, if records do not exist, then the address is unknown. When the Supreme Court created the priority rules, however, it was addressing the escheatment of intangible personal property, such as money orders, where the property clearly existed but the owner’s address could not be ascertained from the four corners of that property or related records. See, e.g., Pennsylvania v. New York, 407 U.S. 206, 207, 92 S.Ct. 2075, 32 L.Ed.2d 693 (1972). Defendants’ logic stretches the definition of address unknown property to troubling lengths.
More important, defendants’ logic is contrary to the fundamental principle of estimation; Defendants are using the existence of unclaimed property in base years to infer the existence of unclaimed property in the reach back years. But defendants are not extrapolating the characteristics of the property that would reduce the liability owed to Delaware. If the property in base years shows an address in another state, then the characteristic of that property has to be extrapolated into the reach back years. Defendants simply did not do that. Because defendants employed estimation in a manner where the characteristics and qualities of the property within the sample were not replicated across the whole, it created significantly misleading results.
5. Multiple Liability
The U.S. Supreme Court held in the Texas cases that “the same property cannot constitutionally be escheated by more than one State.” Texas, 379 U.S. at 678-79, 85 S.Ct. 626. Accordingly, if plaintiff escheats property to Texas, plaintiff cannot escheat that same property to Delaware. If a state forces a holder to escheat the same property already escheated to another state, then the holder is subjected to what plaintiff calls “multiple liability.” (D.I. 113 at 10). No case has addressed whether estimation based on the same property results in multiple liability. But it seems logical that if two states use the same property in the base years to infer the existence of unclaimed property in the reach back years, then a holder is being *550compelled to escheat the same estimated property to two states, in violation of the principles articulated ■ in the Texas cases. This is exactly what happened to plaintiff.
A Texas audit of plaintiff covered the same nineteen years as Delaware’s audit. (D.1.119 at 7-8). Texas used payroll cheeks from 1999-2005 that were payable to -Texas employees to estimate plaintiffs liability for years 1996-1998., (D.I. 113 at 15). Defendants also used payroll checks from 2004-2005 payable to Texas employees to estimate plaintiffs liability for the years 1996-1998. (Id.). As a result, plaintiff estimates that it was subject to at least $299,085.86 in multiple liability.19 (Id.).
■ Defendants assert that there is no risk of multiple liability with the use of estimation, because only the state seeking to escheat property under the secondary rule can use estimation. (D.I. 121 at 20 of 44). In other words, because Delaware is the state of plaintiffs incorporation and seeks to escheat address unknown property, it is the only state that can use estimation to calculate plaintiffs unclaimed property liability. (Id.). As a result, in defendants’ view, other states, like Texas, seeking to escheat under the primary rule cannot use estimation. (Id.). This, however, is neither the law nor the custom'. Indeed, none of the states that have adopted statutes permitting the use of estimation, including Delaware, have expressly limited the use of estimation to the secondary rule, and Texas’ audit of plaintiff is clear evidence that, in practice, states use estimation when calculating liability under the primary rule.
Defendants also assert that there is no risk of multiple liability because the State Escheator is statutorily obligated to indemnify plaintiff against any claims from another person or state that claims the property. (D.I. 115 at 15). Indemnification is not, however, adequate protection. There is no identifiable property in an estimation to which another state could prove it had a priority claim under the primary rule, especially if the other state estimates a liability too. See Texas, 379 U.S. at 682, 85 S.Ct. 626 (another state must “come[] forward with proof that it has a superior right to escheat”).
6. Summary
In summary, defendants: (i) waited 22 years to audit plaintiff; (ii) exploited loopholes in the statute of limitations; (iii) never properly notified holders regarding the need to maintain unclaimed property records longer than is standard; (iv) failed to articulate any legitimate state interest in retroactively applying Section 1155 except to raise revenue; (v) employed a method of estimation where characteristics that favored liability were replicated across the whole, but characteristics that reduced liability were ignored; and (viii) subjected plaintiff to multiple liability. To put the matter gently, defendants have engaged in a game of “gotcha” that shocks the conscience.
Nevertheless, the court is given some pause when contemplating appropriate remedies for defendants’ violation of due process. It is defendants who are best able to know which remedy will be the most palatable in its anticipated efforts to normalize the enforcement of its unclaimed property laws. Thus, the court will defer *551its decision on the subject of an appropriate remedy until another day.
C. The Takings Clause
The takings clause prohibits states from taking private property for public use without just compensation. U.S, Const, amend. V, XIV. To succeed on a takings claim, plaintiff must show that the state’s action affected a “legally cognizable property interest.” Am. Exp. Travel Related Servs., Inc. v. Sidamon-Eristoff, 669 F.3d 359, 371 (3d Cir.2012) (quoting Prometheus Radio Project v. FCC, 373 F.3d 372, 428 (3d Cir.2004)). A holder generally has no property interest in abandoned property. Delaware, 507 U.S. at 502, 113 S.Ct. 1550 (stating that abandoned funds become subject to escheat because the holder has no property interest in the funds). Thus, there is no unlawful taking when a state seeks to escheat abandoned property. Riggs, 581 A.2d at 1243 (noting that a takings claim would fail because the abandoned funds to be escheated to the state were not the property of plaintiff).
The U.S. Supreme Court has also held that the taking clause does not prohibit the mere imposition of an obligation to pay money, such as a tax or penalty, but will prohibit the taking of a specific fund of money. See Koontz v. St. Johns River Water Mgmt. Dist., — U.S.-, 133 S.Ct. 2586, 186 L.Ed.2d 697 (2013) (“It is beyond dispute that taxes and user fees are not takings.” (internal punctuation omitted)); Phillips v. Wash. Legal Found., 524 U.S. 156, 160, 118 S.Ct. 1925, 141 L.Ed.2d 174 (1998) (holding that interest income generated by funds held in IOLTA accounts is private property of the owner of the principal for purposes of the takings clause); Webb’s Fabulous Pharm., Inc. v. Beckwith, 449 U.S. 155, 164-65, 101 S.Ct. 446, 66 L.Ed.2d 358 (1980) (holding that the takings clause can apply to monetary interest generated from the operation of a specific, separately identifiable fund of money). Defendants have repeatedly asserted that Section 1155 is not a penalty.20 (D.I. 25 at 19; D.I. 28 at 18; D.I. 115 at 30, D.I. 121 at 10, 14, 29; D.I. 127 at 4 n. 1). Accordingly, defendants have used Section 1155 to identify a specific fund of money, which, if not unclaimed property, is a legitimate property interest protected by the takings clause.
Unfortunately, the parties have taken an absolutist approach to their argument, with each side claiming that all of the funds are, or are not, unclaimed property. Neither party, however, can be entirely correct. There is a risk that defendants’ method of estimation could be so inaccurate as to sweep into its final assessment property that is not in fact abandoned, but rightfully belongs to plaintiff. Defendants have acknowledged this risk. (See D.1.115-4 at 60 (stating that “[t]he State expects that some portion of a statistical sample will not be unclaimed property”)).
The court has already concluded that the proper use of estimation can satisfy defendants’ burden that property is unclaimed. Estimation is properly employed when it balances the competing interests between an unlawful taking by the state and improper windfall for holders. It appears the Delaware General Assembly tried to strike a balance by requiring that any estimate be “reasonable.” See 12 Del. C. § 1155. There are other contexts where a “reasonableness” standard is used to determine whether an unconstitutional taking of money has occurred. See Duquesne Light Co. v. Barasch, 488 U.S. 299, 307, 109 S.Ct. 609, 102 L.Ed.2d 646 (1989) (explaining that, in the field of utility rate *552regulation-, a reasonable rate is not a taking). Accordingly, the court holds that a reasonable estimation of a holders’ unclaimed property liability is not an unconstitutional taking. The parties have not presented evidence or argument to the court as to whether or not thé estimation was reasonable, although the court is aware that expert reports have been prepared on this issue. Thus, there is a dispute of material fact that cannot be resolved on summary judgment.
D. Prohibition of Ex Post Facto Laws
The law provides a very comprehensive analytical framework for determining when a civil penalty violates the ex post facto clause, which neither party adequately addressed. A law violates the ex post facto clause if it “imposes a punishment for an act which was not punishable at the time it was committed; or imposes additional punishment to that then prescribed.” Coady v. Vaughn, 251 F.3d 480, 487 (3d Cir.2001) (quoting Weaver v. Graham, 450 U.S. 24, 28, 101 S.Ct. 960, 67 L.Ed.2d 17 (1981)). The ex post facto clause is limited, however, to criminal punishments and civil measures which are really criminal punishments in disguise. Riley v. Corbett, 622 Fed.Appx. 93, 95 (3d Cir.2015); Harisiades v. Shaughnessy, 342 U.S. 580, 595, 72 S.Ct. 512, 96 L.Ed. 586 (1952); Eastern Enterprises v. Apfel, 524 U.S. 498, 538, 118 S.Ct. 2131, 141 L.Ed.2d 451 (1998) (noting that for over 200 years the Supreme Court has limited the ex post facto clause to the criminal context).
A statute is a disguised criminal punishment if: (i) the legislature demonstrated an express or implied intent that the statute be a criminal punishment; or (ii) the statute is “so punitive in purpose or effect as to negate the State’s intention to deem it civil.” Smith v. Doe, 538 U.S. 84, 92, 123 S.Ct. 1140, 155 L.Ed.2d 164 (2003); Riley, 622 Fed.Appx. at 95. “[OJnly the clearest proof will suffice to ... transform what has been denominated a civil remedy into a criminal penalty.” Smith, 538 U.S. at 92, 123 S.Ct. 1140. Applying this test here, the court finds that the Delaware General Assembly did not intend Section 1155 to act as a criminal punishment, and the statute does not operate like a criminal punishment.
1. Intent
When it enacted Section 1155, the Delaware General Assembly did not explicitly disclose either a civil or punitive intent. Several facts, however, indicate that the General Assembly intended to implement a civil measure. The statute is codified within title 12 of the Delaware Code, which is a civil, not criminal, title. The only penalty authorized under the UPL is referred to as a “civil penalty.” See 12 Del C. § 1159. And any objections to the audit can be appealed only to the Court of Chancery, which does not have subject matter jurisdiction over criminal offenses. See 12 Del. C. § U56(j) (providing for appeal to Court of Chancery). Other courts have found these facts sufficient to infer a civil intent. See, e.g., Kansas v. Hendricks, 521 U.S. 346, 361, 117 S.Ct. 2072, 138 L.Ed.2d 501 (1997) (finding that state’s intent to create a civil proceeding was evidenced by act’s placement in the civil code); United States v. Harley, 315 Fed.Appx. 437, 440-41 (3d Cir.2009) (finding it probative that the act was codified in the title for public health and welfare, rather than the title for crimes and criminal procedure); United States v. Ward, 448 U.S. 242, 249, 100 S.Ct. 2636, 65 L.Ed.2d 742 (1980) (finding it quite clear that Congress intended to impose a civil penalty when, among other things, it labeled the sanction a “civil penalty”).
2. The Mendoza-Martinez Factors
To determine whether Section 1155 is “so punitive- either in purpose or *553effect” that it should be deemed criminal, the United State Supreme Court has set forth seven factors to consider. Kennedy v. Mendoza-Martinez, 372 U.S. 144, 168-69, 83 S.Ct. 554, 9 L.Ed.2d 644 (1963). The Mendoza-Martinez factors weigh in favor of finding a criminal penalty if: (i) “the sanction involves an affirmative disability or restraint;” (ii) the penalty has “historically been regarded as a punishment;” (iii) the penalty “comes into play only on a finding of scienter;” (iv) the penalty “will promote the traditional aims of punishment—retribution and deterrence;” (v) “the behavior to which [the penalty] applies is already a crime;” (vi) the statute is not rationally connected to an alternative, non-punitive purpose; and (vii) the penalty “appears excessive in relation to the alternative purpose assigned.” Id. Neither party addressed these factors in their briefs, but the court finds that they provide useful guidance in concluding that Section 1155 is not a disguised criminal penalty.21
At least five factors clearly weigh in favor of finding that Section 1155 is- not so punitive as to negate its civil intent. Monetary penalties “do not involve an ‘affirmative disability or restraint.’” Hudson v. United States, 522 U.S. 93, 94, 118 S.Ct. 488, 139 L.Ed.2d 450 (1997); U.S. ex rel. Bergman v. Abbot Labs., 995 F.Supp.2d 357, 384 (E.D.Pa.2014). A monetary penalty has not “historically been viewed as punishment.” Hudson, 522 U.S. at 104, 118 S.Ct. 488; S.E.C. v. Palmisano, 135 F.3d 860, 866 (2d Cir.1998). There is no scienter requirement in Section 1155. See Moyer v. Alameida, 184 Fed.Appx. 633, 639 (9th Cir.2006) (finding that the third factor indicates that the fee is civil, since the statute did not require any scienter). Section 1155 does not apply to behavior that can be punished as a crime. And, there are alternative, non-punitive purposes to which Section 1155 can be rationally connected: to provide for the care and custody • of abandoned property until it can be reclaimed by the true owner. See 12 Del C. § 1144(a).
In addition, there is not clear proof that Section 1155 promotes the traditional aims of punishment. The absence of a scienter requirement is evidence that the statute “is not intended to be retributive.” Hendricks, 521 U.S. at 362, 117 S.Ct. 2072. Moreover, any deterrent effect from the statute’s penalty is not dispositive, because deterrence' “may serve civil as well as criminal goals.” United States v. Ursery, 518 U.S. 267, 292, 116 S.Ct. 2135, 135 L.Ed.2d 549 (1996); Hudson, 522 U.S. at 102, 118 S.Ct. 488 (stating that “all civil penalties have some deterrent effect”).
Plaintiff raises only one argument in support of its assertion that Section 1155 is criminally punitive. According to plaintiff, the use of estimation has resulted in an assessment that is 7,541 times greater than all authorized penalties under the UPL combined. (D.I. 113 at 35). The court will assume for the purposes of its analysis of this factor that Section 1155 is excessive in relation to the alternative purpose of safekeeping unclaimed property.
Overall, however, a balancing of the Mendoza-Martinez factors weighs overwhelmingly in favor of the conclusion that Section 1155 does not have a criminally punitive purpose or effect. Although no one factor is dispositive, the fact that Section 1155 may be excessive in relation to *554its purpose does not provide the clearest proof needed to “transform what has been denominated a civil penalty into a criminal penalty.” Smith, 538 U.S. at 97, 123 S.Ct. 1140. Therefore, the court concludes that, as a matter of law, plaintiff has not stated a claim that defendants’ application of Section 1155 violates the ex post facto clause.
VI. CONCLUSION
For the foregoing reasons, plaintiffs motion for summary judgment (D.1.112) is granted as to the substantive due process claim, and denied as to the takings claim and ex post facto claim. Defendants’ motion for summary judgment (D.I. 114) is denied as to the substantive due process and takings claim, and granted as to the ex post facto claim.
Appendix A
Record Retention Statutes for Unclaimed Property Acts as of 2008
Alabama: Ala. Code § 35-12-90 (2004) (10 years)
Alaska: Alaska Stat. § 34.45.300 (2004) (10 years)
Arizona: Ariz. Rev. Stat. § 44-323 (2000) (5 years)
Arkansas: Ark. Code § 18-28-221 (1999) (10 years)
California: None
Colorado: Colo. Rev. Stat. Ann, § 38-13-124 (1992) (5 years)
Connecticut: None
Delaware: None
Florida: Fla. Stat. Ann. § 717.1311 (1996) (5 years)
Georgia: Ga. Code Ann., § 44-12-228 (1990) (10 years)
Hawaii: Haw. Rev. Stat. § 523A-21 (2008) (10 years)
Idaho: Idaho Code § 14-531 (1997) (7 years)
Illinois: 765 III. Comp. Stat. 1025/11(h)(2) (1993) (5 years)
Indiana: Ind. Code § 32-34-1-43 (2002) (10 years)
Iowa: Iowa Code Ann. § 556.11(8)(a) (2000) (4 years)
Kansas: Kan. Stat. Ann. § 58-3964 (1994) (10 years)
Kentucky: Ky. Rev. Stat. § 393.260 (1942) (15 years)*
Louisiana: La. Rev. Stat. Ann. § 9:173 (1997) (10 years)
Maine: Me. Rev. Stat. Ann. tit. 33, § 1972 (1997) (10 years)
Maryland: Md. Code Ann., Com. Law, § 17-322 (1998) (5 years)*
Massachusetts: Mass. Gen. Laws Ann. ch. 200A, § 7B (1981) (5 years)
Michigan: Mioh. Comp. Laws Ann. § 567.252 (1995) (10 years)
Minnesota: Minn. Stat. Ann. § 345.46 (2005) (10 years)*
Mississippi: None
Missouri: Mo. Ann. Stat. § 447.539(9) (1994) (5 years)
Montana: Mont. Code Ann. § 70-9-821 (1997) (10 years)
Nebraska: Neb. Rev. St. § 69-1322 (1992) (7 years)
Nevada: Nev. Rev. Stat. § 120A.700 (2007) (7 years)
New Hampshire: N.H. Rev. Stat. § 471-C:35 (1986) (10 years)
New Jersey: N.J. Stat. Ann; § 46:30B-95 (1997) (5 years)
New Mexico: N.M. Stat. Ann. § 7-8A-21 (1997) (10 years)
New York: N.Y. Aband. Prop. Law- § 1412-a (1980) (5 years)
North Carolina: N.C. Gen. Stat, Ann. § 116B-73 (1999) (10 years)
*555North Dakota: N.D. Cent. Code § 47-30.1-31 (1985) (10 years)
Ohio: Ohio Rev. Code § 169.03(F)(2) (1984) (5 years)
Oklahoma: Okla. Stat, Am. tit. 60, § 679.1 (1991) (10 years)
Oregon: Or. Rev. Stat, § 98.354 (2001) (3 years)
Pennsylvania: 72 PA. Stat. § 1301.16 (1982) (15 years)
Rhode Island: R.I. Gen. Laws § 33-21.1-31 (1986) (7 years)
South Carolina: S.C. Code Ann. § 27-18-320 (1988) (10 years)
South Dakota: S.D. Codified Laws § 43-41B-32 (1992) (10 years)
Tennessee: Tenn. Code Ann. § 66-29-113 (1991) (10 years)
Texas: Tex. Prop. Code Ann. § 74.103 (1985) (10 years)
Utah: Utah Code Ann. § 67-4a-601 (1995) (5 years)
Vermont: Vt. Stat, . Ann. tit. 27, § 1261 (2005) (10 years)
Virginia: Va Code Ann. § 55-210.24:1 (2000) (5 years)
Washington: Wash. Rev. Code § 63.29.310 (1983) (6 years)
West Virginia: W. Va. Code § 36-8-21 (1997) (10 years)
Wisconsin: Wis. Stat. Ann. § 177.31 (2005) (5 years)- -
Wyoming: Wyo, Stat. Ann. § 34-24-132 (1993) (5 years)
* Denotes a statute of limitations, as opposed to a record retention statute.
. In granting the motion to dismiss, the court found that Texas v. New Jersey, 379 U.S. 674, 85 S.Ct. 626, 13 L.Ed.2d 596 (1965) and its progeny did not apply to disputes between private parties and states. Accordingly, plaintiff did not state a claim for violation of federal common law. A Delaware trial court has since ruled that Texas v. New Jersey does apply to disputes between private parties and the states. See Delaware v. Card Compliant LLC, C.A. No. N13C-06-289 FSS [CCLD] at 6, 21 (Del. Super. Ct. Nov. 23, 2015).
. Plaintiff submitted an appendix that included some, but not all, of the pages from the following report: Dept, of Fin., Office of Mgmt. & Budget, Controller Gen. Office, Delaware’s General Fund Revenue Porfolio: a Report Submitted in Fulfillment of Senate Joint Resolution No. 5144a1 General Assembly (Feb. 2008). The court takes judicial notice of those pages omitted from the filing with the court. The full report is available at http://www. finance, delaware. gov/publications/GP2008. pdf.
. In 2014, that number jumped to around $100 million, because Delaware’s Department of Finance “started doing a better job processing claims.” (A651-52, A723).
. Several factors have contributed to the' un-derreporting by businesses. First, the UPL is contained within Delaware's title governing decedents’ estates and fiduciary relations, because the modem practice of escheatment arose out of laws addressing an intestate' decedent’s estate. (All). In the mid-twentieth century, the escheat laws were expanded to include intangible personal property held by businesses. (Id,). But businesses may not have known to look at the title governing decedents' estates to become aware of their obligations. Second, Delaware advises holders to not file a report if they have no unclaimed property, purportedly because the Division of Revenue "did not want to be flooded with a ton of $0 filings.” (A382; A442; A654). Thus, some percentage of businesses do not file a report because they have no unclaimed property. What percentage remains unknown.
. A question arises as to what role, if any, Delaware’s concern about reduced revenue played in its decision not to adopt a record retention statue. (See DE00001115 (noting that improved record keeping could reduce revenue from unclaimed property)).
. This example uses the numbers from Kel-mar’s report, even though Kelmar’s calculations of net liability for each year is actually a few cents higher than it should be. For exam-pie, $ 3,424,050,000 x 0.00365273% equals $ 125,071.30, not $ 125,071.36, as Kelmar reports. (A462).
.The court has not been asked to determine if the formula used for estimation is reasonable. But Kelmar’s use of “calendar sales” as the denominator does raise questions, because, as the state acknowledged, "abandoned property has no pertinent link to the economy.” (A42-43).
. It is unclear exactly how this extrapolation was done as the unremediated checks represented only 0.07% of the sample set for accounts payable (i.e., $4,508.75 of $6,112,942.10), but Kelmar determined that 0.86% of all aged checks (i.e. $141,189.72 of $16,437,927.15) was unremediated.
. Plaintiffs expert produced several excel spreadsheets illustrating how Kelmar’s calculations would change based on removing certain discrete inputs, but none illustrating this overall point. (See A54-77). It is unknown how much further this number would decrease if the unremediated checks with addresses in other states were also removed from Kelmar’s calculations.
. As part of their argument for Pullman abstention, defendants raise the fact that plaintiff could have appealed directly to the Court of Chancery for a review of the reasonableness of the estimation. (D.I. 115 at 34). This fact, however, is not relevant to determining whether to abstain under Pullman. It might have been relevant to an analysis of whether abstention was ‘ warranted under Burford v. *540Sun Oil Co., 319 U.S. 315, 63 S.Ct. 1098, 87 L.Ed. 1424 (1943), which arises when federal lawsuits would unduly disrupt state administrative efforts to establish a coherent policy with respect to a matter of substantial public concern. Riley v. Simmons, 45 F.3d 764, 771 (3d Cir.1995). But, defendants did not identify or argue Burford, and the court is aware of no obligation to raise and resolve issues of Bur-ford abstention sua sponte. Even if the court was obligated to consider the issue on its own, several facts raise doubts that it is appropriate here. For example, Burford requires timely and adequate state law review to be available. Id. But the deadline for filing an appeal with the Court of Chancery expired almost two years ago, in May 2014. See 12 Del. C. § 1156(j) (requiring appeal to be filed in 30 days). It is also unclear whether the Court of Chancery has authority to consider plaintiff's federal constitutional challenges when Section 1156 expressly limits the court’s review “to whether the independent reviewer’s determination was supported by substantial evidence on the record.” Id. No Delaware case has construed the scope of the court’s authority under Section 1156, perhaps because no parties have yet availed themselves of the process. None of this, of course, forecloses the possibility that defendants may persuasively argue for Burford abstention in future cases.
. The applicable statute read: "When property is paid or delivered to the State Treasurer under this article, the owner is entitled to receive income or other increments actually received by the State Treasurer.” Smolow, 353 F.Supp.2d at 565 n. 1.
. Both parties relegated the issue of laches to a few footnotes, with plaintiff identifying only one case, from Pennsylvania, finding that laches applied to the government. (See D.I. 113 at 22 n. 35; D.I. 119 at 15 h. 18; D.I. 121 at 29 n. 14). The court, however, found Delaware cases recognizing that laches may apply to the state government. See State ex rel. Brady v. Pettinaro Enter., 870 A.2d 513, 528 n. 32 (Del.Ch.2005) (stating that ”[t]he usé of lach-es against the State, while raising unique considerations, is not unprecedented,” and providing examples). Because plaintiff did not seriously pursue any laches argument, the court will assume that the unique circumstances Warranting applicátion of laches against the state are not present here. This does not foreclose the possibility that another plaintiff may persuasively argue a laches defense in future cases. .
. In July 2015, the General Assembly adopted a series of amendments to Section 1155 phasing in how far back defendants’ audit can reach when not limited by the statute of limitations. See S.B. 141, 148th Gen. Assembly (2015). Effective January T, 2017, the State Escheator cannot1 seek a deficiency payment for any transaction that is more than 22 years old. 12 Del. C. § 1155(e). Because these amendments were adopted after plaintiff initiated this action, they are not applicable to this case. Nevertheless, the amendments would have had no impact, because, conveniently,, defendants' audit of plaintiff reached back only 22 years,
. Plaintiff was able to reproduce the unclaimed property1 reports it filed in 1996, 1998, and 1999. (A462). Defendants assessed ' a liability for those years. It is not clear why a ■ -liability for those years is not; barred by the statute of limitations.
. In Delaware, the U.S. Supreme Court'used the terms "debtor” and "creditor” in place of "holder” and “owner.” Delaware, 507 U.S. at 499, 113 S.Ct. 1550.
. Defendants also claim that they are not applying Section 1155 retroactively, because holders have been required to deliver unclaimed property to the state since 1971. (D.I. 115 at 17-18; D.I. 127 at 6). This argument fails because the Supreme Court has recognized that amendments increasing monetary liability for same conduct operate retroactively. See, e.g., Hughes Aircraft Co. v. U.S. ex rel. Schumer, 520 U.S. 939, 947, 117 S.Ct. 1871, 138 L.Ed.2d 135 (1997) (finding that an amendment increasing monetary liability, but not altering the normative scope of unlawful activity, was retroactive); Rivers v. Roadway Express, Inc., 511 U.S. 298, 303, 114 S.Ct. 1510, 128 L.Ed.2d 274 (1994) (holding that an increase in monetary liability is retroactive even though the normative scope of law was not altered). Section 1155 increases monetary liability for the same conduct and, therefore, it operates retroactively,
. To illustrate this point, consider the fact that courts have upheld, in the face of substantive due process challenges, the unlimited retroactivé application of the Comprehensive Environmental Response Compensation and Liability Act. See, e.g., United States v. Monsanto Co., 858 F.2d 160, 173-74 (4th Cir. 1988).
. Owners are the appropriate public constituency to consider in a burden shifting analysis. The public is not the appropriate constituency, because then the state would be raising revenue for the general welfare, which is essentially how a tax operates. Although the state can use unclaimed property for the general welfare, unclaimed property is not to be treated like a tax. Defendants admit that "[a]bandoned property is not a tax.” (A588).
. Defendants eventually gave plaintiff a credit of $100,914.92 based on the amount plaintiff actually escheated to Texas for the years of 1996-1998. But this credit did not eliminate the multiple liability issue, because defendants used the Texas payroll checks to estimate that plaintiff owed Delaware $299,085.86 for the years of 1996-998. (D.I. 115 at 9 n.2). The multiple liability issue would be eliminated only if defendants credited plaintiff the full amount it derived from the use of the Texas payroll checks.
. Defendants cannot claim that they were only responding to plaintiff’s ex post facto claim when they argued that Section 1155 was not a penalty. Defendants did not limit their assertion to only the ex post facto context. (See, e.g., D.I. 121 at 10, 14, 22).
. Defendants did not identify or discuss the Mendoza-Martinez factors in either this motion or its earlier motion to dismiss. (See D.I. 115 at 29-31). It is surprising that defendants did not comprehensively address the entire analytical framework for determining when a civil penalty is a disguised criminal penalty. {See D.I. 25 at 17-19). If defendants had brought to the, court’s attention the complete analytical framework for an ex post facto claim, it is unlikely the claim would' have survived defendants’ motion to dismiss.