Economic, Legal and Social Dimensions of Regulatory Arbitrage

1. Introduction – The Economic, Legal and Social Dimension of Regulatory Arbitrage, by Jan Friedrich and Matthias Thiemann, https://doi.org/10.1515/ael-2020-0150.

2. Regulatory Arbitrage – What’s Law Got To Do With It?, by Katja Langenbucher, https://doi.org/10.1515/ael-2019-0054.

3. Regulatory Arbitrage and Non-Judicial Debt Collection in Central and Eastern Europe – Tax Sheltering and Potential Money Laundering, by Cătălin-Gabriel Stănescu and Camelia Bogdan, https://doi.org/10.1515/ael-2019-0055.

4. Ambiguities in Accounting and their Impact on Regulatory Arbitrage: A Study on the Anchoring of the Rights and Obligations Approach in the IASB’s Conceptual Framework, by Tessa Kunkel, https://doi.org/10.1515/ael-2019-0049.

5. Regulatory Arbitrage in the Intersection of Accounting Standards and Tax Laws: The Case of Synthetic Leases, by Jan Friedrich, https://doi.org/10.1515/ael-2020-0098.

6. Detecting Tail Risks to Preclude Regulatory Arbitrage – The Case for a Normatively Charged Approach to Regulating Shadow Banking, by Matthias Thiemann and Tobias Tröger, https://doi.org/10.1515/ael-2019-0059.

1 Introduction

The off-balance sheet treatment of leases has been criticized for decades by academics (Briloff, 1976; Donegan & Sunder, 1989; Myers, 1964), prudential regulators (ESMA, 2013; IOSCO, 2011; SEC, 1949, 2003a, b)^[1] and accounting standard setters (Herz, 2016), as well as by those who prominently used sophisticated off-balance sheet structures to manipulate the financial statements of their corporations (see A. Fastow, former CFO of Enron, in BBC, 2015).^[2]

As part of the convergence project with the Financial Accounting Standards Board (FASB) and after years of controversial debates, the International Accounting Standards Board (IASB) replaced the prior IAS 17 and issued, in January 2016, IFRS 16 ‘Leases’ that should bring essentially all leases onto the lessee’s balance sheet. Only one month later, the FASB released the counterpart to IFRS 16, the ASC 842, replacing SFAS 13 (ASC 840). In particular, “[t]he synthetic lease has been the poster child of the movement to reform lease accounting” (Bosco, 2011, p. 1), as Enron and other stars of the so called ‘New Economy’ used this off-balance sheet tool to hide an essential part of their debt exposure, enhancing their leverage and deferring cash outflows while obtaining tax benefits (Baker & Hayes, 2004; Dharan, 2002; Duke & Hsieh, 2006; Newman, 2007). A synthetic lease is a hybrid between an off-balance sheet operating lease and on-balance debt financing, blurring the line of ownership within a single transaction by means of regulatory arbitrage techniques, as depicted by Figure 1 below:

Figure 1:

Synthetic Leases.

In a synthetic lease, the lessee is treated as the owner of the leased asset for tax purposes and, at the same time, not as the owner of the asset for financial reporting purposes, as managers seek “to get the best of both worlds” (Hanlon & Heitzman, 2010, p. 133). Although a synthetic lease “is a pure credit transaction” (Mayer, 2005, p. 13), managers engage in cross-regulatory arbitrage (Huang, 2009) between tax laws and accounting standards as the lease remains off balance sheet for the lessee’s financial reporting, while depreciations and interest expenses can be deducted for tax purposes by that same lessee. The revised lease accounting standard intentionally avoids the definition of clear bright lines as precise accounting standards have been criticized for facilitating structuring activities (Nelson, 2003; Nelson, Elliott, & Tarpley, 2002). The revised IFRS 16 and ASC 842 widely refrain from the severely criticized threshold approach and moves towards a continuous accounting approach, which has been hailed to limit regulatory arbitrage (Dye, Glover, & Sunder, 2015). In a continuous approach, a “contract set would specify the rights and obligations of each agent under all contingencies” (Donegan & Sunder, 1989, p. 212), depending on the entire probability distribution of the potential obligation. Hence, proponents state that “small manipulations can have only small consequences” (Dye et al., 2015, p. 292, footnote 44) as the present value of minimum lease obligations appears on the lessee’s balance sheet regardless of other contractual features of the lease” (ibid., p. 291).

However, despite the recent revision of lease accounting standards, synthetic leases could be restructured in the USA in response to the counterpart of IFRS 16 under US-GAAP – ASC 842 – in order to maintain a substantial proportion of the lease exposure off balance sheet (Biondi et al., 2011; Bosco, 2019; ELFA, 2017; see also Kakoschke, 2004 for synthetic leases in Canada; see Huang, 2009 for Australia). Furthermore, the “Tax Cuts and Jobs Act” from 2017 even seems to fuel the growth of synthetic leases in the USA once again (Borders et al., 2018; Kessler, 2019), and “it seems ironic that synthetic-like structures are likely to fare the best of all operating lease structures” (Bosco, 2011, p. 3) under the new lease accounting standards.^[3]

This raises important questions regarding the reciprocal linkages between accounting standards and tax laws, as synthetic leases are subject to political economy in both regulatory areas. Research on regulatory arbitrage focuses on either tax avoidance (Avi-Yonah & Pichhadze, 2017; Buettner & Thiemann, 2017; Fleischer, 2010) or accounting gimmicks (Dye et al., 2015; Donegan & Sunder, 1989; Thiemann, 2018). However, the case of synthetic leases shows that tax laws and accounting standards might be reciprocally linked in certain transactions (Biondi, 2017; Schmiel & Weitz, 2017; Sunder, 2011), creating a legal web of overlapping regulations (Biondi, 2019; Langenbucher, 2020 in this issue). Hence, this paper argues that these financial tools can only be fully understood if we think of both accounting and tax arrays together.

Research on synthetic leases has primarily focused on the economic consequences of recognition (Altamuro, 2006; Callahan, Smith, & Spencer, 2013; Zechman, 2010; Zhang, 2009). However, little is known about the political economy of financial tools that are driven by regulatory arbitrage in the intersection of tax laws and accounting standards. Examining the “cat-and-mouse game between rule-makers and preparers” (Dye et al., 2015, p. 266) for the case of synthetic leases in the USA, this paper shows that the development of this sophisticated off-balance sheet instrument has to be understood as a co-evolution of tax laws and accounting standards. Besides academic sources, the paper essentially draws upon documents (including blogs and presentations at industry events) from the leasing industry, specialized lawyers or law firms, auditing firms, and financial engineers to get a fine-grained understanding of regulatory arbitrage techniques in the structuring of synthetic leases. Analyzing the perpetual cycle of rule setting, rule evasion, and re-regulation (Kane, 1988) as a multi-level game, this paper seeks to make the following contributions.

Firstly, the paper engages in the discussion over principles- and rules-based accounting standards (Maines et al., 2003b; Nelson, 2003; Nobes, 2005; Wuestemann & Wuestemann, 2010) and tax laws (Avi-Yonah & Pichhadze, 2017; Kaplow, 1992; Weisbach, 1999). The paper contributes to this debate by analyzing the complementary interplay between accounting standards and tax laws as a co-evolution (Sawabe, 2005). More specifically, it shows that efforts, since the early 2000s, towards more principles-based accounting standards through convergence of national accounting standards with IFRS (c.f. Camfferman and Zeff, 2015) may create new structuring opportunities if these standards are combined with rules-based tax legislation.

Secondly, this article contributes to the literature on book-tax-conformity (Desai, 2005; Hanlon & Maydew, 2009; Manzon & Plesko, 2002; Mills & Newberry, 2005; Schmiel & Weitz, 2017; Sunder, 2011) which focuses on the magnitude, sources and effects of diverging financial and tax reporting. While research in this area primarily focuses on the gap between reported and tax income, little is known about the relationship between tax optimization and off-balance sheet financing. Exploring the evolution of synthetic leases, the paper shows that off-balance sheet financing may be provoked by the attempts of lessees to obtain tax benefits and vice versa.

Thirdly, the paper contributes to the discussion over a re-examination of the accounting consensus (Sunder, 2009), as “[t]he pendulum appears to have swung too far in the direction of uniform written standards” (ibid., p. 109), neglecting the role of historically grown norms and institutions (Sunder, 2002, 2005). While being necessarily more rules-oriented, tax legislation institutionalizes at least some parts of these social norms. This paper argues that globally uniform accounting standards can be usefully combined with diverging institutional set-ups by means of selective conformity (Luppino, 2003; Raby & Richter, 1975) between accounting standards and tax laws. In contrast to a universal conformity of financial and tax reporting (i.e., book-tax conformity), national state actors or agencies may impose conformity on a selective basis for those financial innovations, which create a regulatory benefit from a diverging accounting and tax treatment.^[4] This solution would avoid the shortcomings of a one-book system (Hanlon, Maydew, & Shevlin, 2008; Shaviro, 2008). Hence, the paper proposes such a selective conformity as an additional means of national actors to quickly react to newly emerging financial innovations, involving regulatory arbitrage, in order to mitigate unintended consequences of international accounting convergence. This approach might be fruitfully combined with other innovative regulatory set-ups such as an ‘embedded’ market regulator (Thiemann & Lepoutre, 2017; Thiemann & Troeger, 2020 in this issue), as well as with retroactive taxation (Langenbucher, 2020) on off-balance-sheet instruments that are identified by securities regulators, tax authorities or accounting standard setters for having no value in use, besides the evasion of regulatory requirements.

The rest of the paper is organized as follows: in the following section, a literature review on the role of accounting and taxation is given and discussed in the context of regulatory arbitrage. The Section 3 briefly introduces the basic mechanics of synthetic leases. Following that, the perpetual evolution of synthetic leases in the USA is explored, covering the time period from the early 1990s to 2020. The paper concludes with a discussion of the findings and potential policy implications.

2 Regulatory Arbitrage in the Realm of Accounting and Taxation

Accounting research on regulatory arbitrage has pointed to the tense relationship between the regulatory intent of a norm and the written words (Gilson, 1984; Kane, 1988; Shah, 1996; Sunder, 2009; Thiemann, 2018), as “regulatory arbitrage exploits the gap between the economic substance of a transaction and its legal or regulatory treatment” (Fleischer, 2010, p. 3). By taking advantage of the inability of written letters to precisely meet the spirit of the law (ibid.), “market actors use law, in the form of legal instruments, to avoid law, in the form of legal restrictions” (Black, 2013, p. 21). In this way, regulatory arbitrage techniques seek to game with regulatory burdens by creating a beneficial regulatory treatment while undermining the spirit of a regulatory norm (Fleischer, 2010, p. 3). In this game, the tense relationship between legal certainty and uncertainty is crucial. Thiemann and Lepoutre (2017, p. 37) argue that “[p]rinciples-based regulation […] provides only limited ex ante certainty about the regulatory classification”, while rules-based systems “made the work of creative compliance easier for market actors since legal engineers could easily avoid binding criteria” (ibid., p. 19). On the other hand, James (2010, p. 581) states that “uncertainty also arises in rules-based legislation because the interpretation of legislation is a matter of the interpretation of words and uncertainty is a result of instances where their meaning in a particular context is considered to be ambiguous”.

Accounting scholars have stressed for a long time that precise bright lines facilitate regulatory arbitrage and, as a consequence, called to shift from rules-based towards more principles-based accounting standards (Dye et al., 2015; Maines et al., 2003b; Nelson, 2003; Nobes, 2005; comp. Gill, 2002 and Schipper, 2003 for a critical stance). More specifically, Nelson et al. (2002) and Agoglia, Doupnik, and Tsakumis (2011) find that precise accounting standards make structuring activities by managers more likely, while these attempts are less likely to be challenged by auditors. Additionally, Kang and Lin (2011) have shown that managers with motivation for aggressive reporting are more likely to do so under precise accounting standards.

The precise formulations of most lease accounting standards around the world have been strongly criticized, as they typically follow an “all-or-nothing” (Dworkin, 1967, p. 25) approach. This involves a dichotomy choice between one kind of accounting treatment or another, according to some specifications. For instance, concerning leases, it requires choosing one type or another if a certain quantitative threshold is passed. This means that one lease at 79.99% will be treated in one way, while another one at 80.01% will be treated in a completely different way for sake of financial accounting and reporting. Such dichotomy of when applied to lease accounting standards is prone to regulatory arbitrage “because small changes in the terms of contracts in the neighborhood threshold values can have large effects on the appearance of financial statements” (Donegan & Sunder, 1989, p. 216). Imhoff and Thomas (1988) have shown that corporations exploited the quantitative thresholds of SFAS 13^[5] under US-GAAP and engaged in regulatory arbitrage techniques in order to avoid the recognition of lease obligations on the lessee’s balance sheet. Collins, Pasewark, and Riley (2012) compare the lease exposure of firms (similar size and operating in the same industry) using the more rules-based lease accounting standards under US-GAAP with firms that prepare their financial statements in accordance with the more principles-based IAS 17 under IFRS. Their study finds that less precise lease accounting causes a higher degree of recognition. The experimental work of Cohen, Krishnamoorthy, Peytcheva, and Wright (2013) supports these results, showing that more principles-based lease accounting standards result in less aggressive financial reporting, particularly in combination with a weak level of enforcement by regulatory agencies (see also Abdel-Khalik, 2019 on the role of SEC in the Enron debacle). Additionally, Mills and Newberry (2005) find that firms with a weaker financial standing are more likely to use structured off-balance sheet tools, such as synthetic leases. The study suggests that this is particularly true when corporations are at risk of violating debt covenant arrangements (ibid., p. 279).

However, the former chair of the FASB rightly insists that “just writing more principles-based or objectives-oriented standards does not ensure that the result will be improved, conceptually consistent, and principles-based financial reporting” (Herz, 2016, p. 297) as the accounting standards in the USA “are part of the financial reporting supply chain” (ibid.), involving various actors beyond accountants. For instance, financial intermediaries are crucial for the development of off-balance sheet financial innovations (see Funk & Hirschman, 2014 on OTC Swap derivatives) or for the structuring of securitization arrangements (Engelen et al., 2010, Thiemann, 2018). As Moreover, as the implementation of off-balance sheet schemes, involving regulatory arbitrage, requires approval by the auditor of a firm, these gatekeepers (Coffee, 2006) form another essential part of the accounting supply chain. This is all the more important as more principles-based accounting standards increasingly delegates interpretative power to preparers and auditors. Using an experimental design, Hackenbrack and Nelson (1996) as well as Kadous, Kennedy, and Peecher (2003) demonstrate that auditors have the propensity to approve incentive-compatible aggressive reporting decisions by exploiting the vague formulation of accounting standards. Backof, Bamber, and Carpenter (2013) have shown that rules-based accounting standards, accompanied by auditor judgment frameworks, significantly improve audit quality, while Agoglia et al. (2011) find that chief financial officers (CFOs) are less likely to engage in aggressive reporting if precise standards are combined with a strong audit committee. Focusing on the auditor’s dependency, Bratton (2004) argues that captured auditors will be disabled to challenge attempts of aggressive reporting under principles-based accounting standards. Hence “[b]roadbrush reformulations of rules-based GAAP should follow only when institutional reforms [that address this issue] have succeeded” (ibid., p. 13). In this vein, Schipper (2003) critically insists that more principles-based accounting standards may reduce the comparability of financial statements, as “[a] general principle is concise and calls for judgment in its application, which must necessarily vary across individuals and situations” (Sunder, 2009, p. 104). All in all, these studies reveal a tense relationship between rules-based and principles-based accounting standards in terms of dealing with regulatory arbitrage. In this game, states and state agencies may play a crucial role in the accounting supply chain, as accounting policy is a central feature of taxation, making both accounting standard settings subject to political economy (Cooper & Sherer, 1984; Friedrich, 2020; Thiemann, 2014, 2018) while creating a web of overlapping requirements (Biondi, 2019; Langenbucher, 2020). Although the international financial reporting standards setter does not recognize this relationship, accounting and taxation are intensively and structurally connected to each other (see Tsunogaya, Okada, & Patel, 2011 on the effect of the Corporation Tax Law on accounting practice in Japan).

In contrast to financial statements that should inform shareholders and creditors about the economic situation of firms, tax laws are designed to serve their own objectives.^[6] Despite the differences, the struggle between principles-based and rules-based tax legislation is an important battleground in the area of taxation as well. Legislation seeks to treat all citizens equally (Du Gay, 2005), avoiding any random error (Kaplow, 1992, p. 594), while individual case-by-case reviews seem to be too costly (ibid., p. 566); tax laws are thus typically rules based. Weisbach (1999, p. 866) stresses that under rules-based tax legislations, “the content of the law is determined before individuals act”. Due to the precise thresholds and bright lines, “taxpayers have been able to manipulate the rules endlessly to produce results clearly not intended by the drafters” (ibid., p. 860).^[7]

Examining three prominent cases of regulatory arbitrage in the USA, Desai (2005, p. 171) argues that “the drive to improve reported book profits fosters tax avoidance and how the drive to limit taxes gives rise to the manipulation of accounting profits and managerial malfeasance”. In other words, tax avoidance and the gaming of reported profits are reciprocally linked. Using a wide range of measures, several studies provide empirical evidence for the significantly increasing gap between reported income and taxable income in the USA since the 1990s (Hanlon & Maydew, 2009; Hanlon & Shevlin, 2005; Mazon & Plesko, 2002; Mills, Newberry, & Trautman, 2002; Yin, 2003). Several scholars suggested an increasing conformity of financial and tax reporting to counter aggressive tax planning (Desai, 2005; Shaviro, 2008; Sunder, 2011; Whitaker, 2005; see Schmiel & Weitz, 2017 for the European Union). However, empirical studies indicate that a broad conformity would deter the value relevance of accounting earnings (Hanlon et al., 2008). Against this background, some scholars suggested the introduction of a selective conformity between tax and accounting representations for certain business transactions (Luppino, 2003; Raby & Richer, 1975). Such selective conformity fosters book-tax conformity only where it is demanded, mitigating the shortcomings of a universal conformity. Notwithstanding whether book financial reporting should converge to tax accounting (Luppino, 2003) or the other way around (Raby & Richer, 1975), proponents of a selective conformity call for an inter-agency coordination between the several parts of the accounting supply chain.

While literature on book-tax conformity primarily focuses on the gap between book and tax income, Hanlon and Heitzman (2010, p. 155) state that “understanding to what extent the off-balance sheet financing affects the analysis of taxes on the choice between debt and equity is an open question”. Examining the development of synthetic leases in the USA as a co-evolution of tax laws and accounting standards, the Sections 3, 4 seeks to shed some light on this issue.

3 The Basic Mechanics of Synthetic Leases

The tax and accounting treatment of off-balance sheet operating and on-balance sheet finance leases in the USA is typically consistently designed “[w]ith the notable exception of synthetic leases” (Mills & Newberry, 2005, p. 255). “This financial instrument stretches the boundaries of property and non-property, of ownership and non-ownership, assets and non-assets, and the definition of conglomerate itself, to the limit by linking the respective manifestation to the addressees of the conglomerate’s financial statement” (Thiemann & Friedrich, 2016, p. 10). Just as the accounting treatment and the tax treatment are disentangled in a synthetic lease, this financing scheme essentially requires a low level of book-tax conformity. The underlying asset typically has a very long economic lifetime (Brightham & Ehrhardt, 2007, p. 717) – such as real estate or railroads – as only in these cases do the tax benefits outweigh the significant structuring costs. The creation of a synthetic lease usually involves coordinated efforts by specialized lawyers, attorneys (bank-related) leasing professionals, tax practitioners, auditors, as well as the lessee’s accountants (Murray, 1997; Robinson & Gough, 1994, p. 23). The structuring is then “encouraged, approved, or at the very least acquiesced in, by firm leadership, which often devoted significant organizational resources to the effort” (Rostain & Regan, 2014, p. 326). Although the tax treatment might be challenged by the tax authority or by courts, “this risk should not prevent well advised parties from proceeding with these transactions once the risk is quantified” (Robinson & Gough, 1994, p. 20). The specific design of synthetic leases evolved over time and differs across jurisdictions and even across federal states. However, synthetic leases also show a set of common characteristics.

In a synthetic lease, the lessee commits to serving fixed payments to the lessor, while the contract typically includes a purchase option at a negotiated price, which enables the lessee to benefit from any appreciations at the end of the lease term (Borders et al., 2018, p. 3). As the amortization for accounting purposes is lower than for tax purposes (or the lease is not amortized at all), synthetic leases defer the cash outflow for the leased asset, allowing the lessee to conserve cash for alternative uses (including distributions to shareholders). The financing rates for synthetic leases are typically similar to the borrowing rates of the lessee’s credit facility (Borders et al., 2018, p. 3) and seem to offer lower funding costs than secured real estate loans (Altamuro, 2006). In addition to the off-balance sheet effect, synthetic leases improve important financial ratios for lessees such as the return on equity (RoE) or the return on asset (RoA) (see Table A1 in the Appendix for a sample calculation under ASC 842). Synthetic leases usually further include a series of renewals. This series allowed an evergreen (perpetual extension) off-balance sheet structure under the previous regulatory treatment (Sebik, 2019) while reducing the upfront recognition under the new ASC 842 or IFRS 16 (if the base lease term – that is, the first one of the series – is shorter than the ‘real’ lease term – situated at the end of the whole series of renewals). The involved Special Purpose Entity (SPE) is typically structured as a bankruptcy-remote entity that “is legally independent of and removed from the bankruptcy risks of the lessee/corporate user” (Murray, 1997, p. 204). For that purpose, financial engineers avoid the classification as secured interest. Therefore, the transaction is treated as a secured loan from the lessor (SPE) to the lessee by failing the requirements of the Uniform Commercial Code (UCC) 2A and 1-201(37), as well as the corresponding regulations (Mayer, 2005). This structural isolation enables the vehicle to issue securitized notes and to lever the investment as a classification as secured loan protects investors and creditors against involuntary claims by the lessee against the SPE if the first approaches insolvency (Murray, 1997, p. 205, 207; Weidner, 2000, p. 450). These leveraged synthetic leases involve a third-party transaction with an unrelated creditor, while the access to capital markets decreases the funding costs of synthetic lease transactions (Murray, 1997, pp. 204–205). With the mandatory application of ASC 842 in 1 January 2021, however, leveraged leases seem to disappear as the new standard does not allow the SPE to offset the debt against the corresponding lease receivables. Notwithstanding a grandfathering of existing leveraged leases, ASC 842 requires the SPE to show the full amount of the asset as well as the corresponding debt, eliminating the positive effects on the RoA of the lessee (Bosco, 2015; see Epstein, 2009, p. 862 on leveraged leases under the previous accounting standard).

In the aftermath of the Enron disaster in 2001, credit agencies state that they treat synthetic leases as financing, while lenders included this financing scheme in their debt covenant agreements (Borders et al., 2018, p. 2; Murray, 2002). However, lessees still seek to obtain an off-balance sheet effect.

4 Synthetic Leases – A Co-Evolution of Accounting Standards and Tax Law

Synthetic leases are part of the evolving movement towards structured finance and securitization in the USA (Dharan, 2002; Soroosh & Ciesielski, 2004). In this vein, synthetic leases also have to be understood in the context of financialization (Bryer, 2004), as the structuring of off-balance sheet SPEs is linked to the increasing focus on capital markets (Dharan, 2002) and essentially involves financial institutions (Abdel-Khalik, 2019; Powers, Troubh, & Winokur, 2002). Synthetic leases occurred together with other innovative off-balance sheet structures as the result of several institutional and regulatory changes. Corporations came under pressure to demonstrate strong equity-to-debt ratios, combined with a progressive dissemination of the shareholder value approach, while precise legal, tax laws and accounting standards were issued, which eased the implementation of off-balance sheet structures (Nesvold, 1998, p. 84, footnote 2). Additionally, the movement in the USA towards book-tax conformity^[8] was stopped in the late 1970s. On the one hand, two crucial actors – the American Institute of Certified Public Accountants (AICPA) and the Internal Revenue Service (IRS) – increasingly fought for competences, as each of them sought to affect the rule setting of the other (Raby & Richer, 1975). Finally, the IRS abandoned the idea of further convergence (Porcano & Tran, 1998, p. 437). On the other hand, two decisions by the Court of Claims in 1976, and particularly by the Supreme Court in 1979, strongly rejected an alignment of reported and taxable income and, in essence, buried the movement towards tax-book conformity in the USA (ibid., pp. 437–438). As a consequence of these institutional and regulatory changes, the gap between reported and taxable income in the USA started to increase, as Figure 2 shows below:

Figure 2:

Book Income versus Taxable Income for Corporations with more than $1 Billion in Assets (in Billions of Dollars).

Source: U.S. Treasury Department (1999).

As part of the general development towards “creative financing” (Weidner, 2000, p. 447), in which “[t]ransactions have been structured to misinform” (Gill, 2002, p. 981), synthetic leases initially occurred in the early 1990s. Synthetic leases became more and more popular in the late 1990s when the fast growing stars of the American New Economy – including the infamous Enron – increasingly applied this off-balance sheet financing scheme to conceal their real debt exposure (Brightham & Ehrhardt, 2007, p. 717). The following sections analyze how financial engineers have been able to perpetually exploit the space between tax and accounting in order to create evolving synthetic lease structures in response to changing regulatory requirements.

4.1 The First Generation of Synthetic Leases (1990–2002)

The synthetic leases of the first generation enabled creditworthy lessees to fully finance the cost of the leased property, while the lease exposure remained off balance sheet (Epstein, 2009, p. 696). The base lease term is typically between five and seven years and can be renewed at mutually agreed conditions (ELA, 2003a, p. 4). In order to create the desired off-balance sheet effect, financial engineers had to deal with a huge number of rules-based regulations, laws and accounting standards that evolved over time.^[9] In particular, synthetic leases exploited the bright-line test under the SFAS 13, as well as the rules-based tax regulations from the Internal Revenue Service (IRS), typically involving the use of an SPE.

In the first step, the lessor establishes an SPE. In a leveraged synthetic lease, this shell company mediates between the lessee, as well as the lessor and the lenders. The SPE serves as a lessor, takes the leased asset, collects the rents from the lessee and services the loan from the lenders with these lease payments (Soroosh & Ciesielski, 2004, p. 31). Due to the 3% bright line of the accounting rule EITF 90-15, the SPE remains off balance sheet for the sponsor (the lessee) as long as a third party (an equity investor) guarantees at least 3% of the equity of the trust.^[10] For the remaining 97% cost of the asset, the SPE obtains a non-recourse loan from at least one financial institution. The SPE acquires an asset and enters into a lease contract with the lessee (Brightham & Ehrhardt, 2007, p. 717; Little, 1997, p. 24).^[11]

In the next step, financial engineers have to ensure that the SPE does not meet all the following recognition criteria under SFAS 13: a) The lease transfers ownership of the property to the lessee by the end of the lease term; b) The lease contains a bargain purchase option; c) The lease term is equal to 75% or more of the estimated economic life of the leased property; d) The present value (PV) is at the beginning of the lease term of the minimum lease payments ≥90% of the fair value of the leased property (SFAS 13, paragraph 7). For example, the bright line under d) can be easily avoided by contractually determining the PV of the minimum lease payments at 89.9% (ELA, 2002, p. 7). In this way, the lease qualifies for an off-balance sheet operating lease under US-GAAP.

Finally, synthetic leases had to be designed to meet the criteria of an on-balance-sheet conditional sale (or secured loan) under tax laws. The Revenue Ruling 55-540, 1955-2 CB 39 Sec(s) expressed the position of the IRS.^[12] The rule set incorporates various court rulings (Murray, 1997, pp. 197–198, footnote 4) and mainly draws upon the intent of the involved parties to determine whether the lease qualifies as either a conditional sale or a lease (Rev. Rul. 55-540, §4). The guidance lists a number of indicators to identify the intent of the contractual parties, and the transaction has to meet at least one of the criteria to qualify for a conditional sale (ibid. §4, 1). Although the IRS stresses that these indicators are not exclusive and are subject to a case-by-case review, “the courts and the Service have applied a fact-based analysis to determine whether the substance of the transaction is in accord with its form and the express intent of the parties” (Murray, 1997, p. 197, footnote 4). The courts primarily focus on the legal form of a transaction and, in a next step, examine whether the contractual agreement is consistent with the economic substance (Newman, 2007, pp. 105–106). Courts identify the economic substance of a transaction by examining whether the lessee or the lessor is vested with the true risks and rewards. For tax purposes, the synthetic lease, hence, is structured to indicate the intent of both parties that the benefits and burdens of ownership reside with the lessee (ibid., p. 106). Robinson and Gough (1994, p. 21) have stressed that “[t]he receipt of a meaningful residual value has been cited by several courts and commentators as the key factor in determining the ownership of property”. For that purpose, the residual guarantee is structured in a way that the lessee maintains the any benefits from the upside (appreciation benefits) as well as the majority of the downside (risk of loss) in the leased asset (ibid., p. 23) and meets the requirements of a secured loan. In order to achieve the desired accounting and tax outcome, the PV of the residual value is designed to remain below the 90% bright line of SFAS 13 in order to keep the lease off-balance sheet, while the appreciation potential mainly resides with the lessee, providing sufficient evidence for tax authorities and courts to treat the lessee as the owner of the asset for tax purposes.^[13] In this way, the rules-based character of the relevant accounting standards and of the related tax regulations enables specialized tax lawyers to disentangle the accounting and tax treatments in synthetic lease transactions in order to create the desired outcome with certainty. Figure 3 below illustrates the basic mechanics of synthetic leases.

Figure 3:

The Basic Structure of Synthetic Leases (First Generation).

The rules-based accounting standards and tax laws enabled financial engineers to precisely draw the thin line between off-balance sheet accounting and an on-balance sheet tax treatment. The precise bright lines for the accounting of SPEs eliminated any uncertainties for preparers and limited regulatory costs, incentivizing managers to create even more off-balance sheet financing vehicles. Using a random sample of U.S. corporations, Soroosh and Ciesielski (2004) demonstrate that the use of SPE-based off-balance sheet structures has increased eightfold from 1999 to 2001.

However, this development was about to change with the collapse of the new economy in 2001 and the downfall of Enron, which was caused by accounting fraud involving a web of SPE-based off-balance sheet schemes. The following revision of accounting standards was intended at bringing structured off-balance sheet vehicles onto the books of preparers, but financial engineers adjusted to the new requirements and synthetic lease structures only evolved.

4.3 The Third Generation of Synthetic Leases (2016–today)

The issuance of IFRS 16 in January 2016 and the publication of the American version ASC 842 one month later fundamentally changed the accounting of lease agreements. Drawing upon the rights-and-obligations approach (see Kunkel, 2020 in this volume), the new standards obliged the lessee to recognize a right-of-use asset, as well as the corresponding lease obligation. After years of intense discussions under a joint effort, the FASB and the IASB ended up with similar lease accounting standards.^[16] However, the new standards differ in some aspects that are crucial for synthetic leases.

At first glance, ASC 842, which became effective in 2019 for listed companies in the USA, seems to be more restrictive than the IFRS 16. The American leasing standard does not exempt low-value assets (less than 5000 USD) that may provide – in addition to the contested bright line to off-balance sheet service contracts – structuring opportunities under IFRS (Thiemann & Friedrich, 2016). The FASB maintained, on the other hand, the distinction between operating and finance leases under ASC 842 in order to mitigate the front-loading effect of the new standard in the lessee’s income statement (PwC, 2016). Although ASC 842 moves towards a more principles-based accounting for leases, the dual-model includes a bright-line test to differentiate between operating and finance leases. Although these bright lines are downgraded to guidelines, which should require further interpretations by the auditor, “at least one of the big four accounting firms has confirmed […] that they will not object to the use of the bright-line tests from ASC 840” (Borders et al., 2018, p. 3).^[17] Although both lease types require the recognition of the lease, they differ in the respective balance-sheet treatment. Most importantly, the classification as operating leases allows synthetic lease structures – in contrast to users of IFRS, including in in the USA – to not be treated as a non-debt obligation under US-GAAP (ELFA, 2016, p. 5).^[18]

Both lease accounting standards, on the other hand, require that the lessee only has to recognize the value of the residual guarantee and not the entire guarantee. In other words, the lessee only has to recognize the PV of the lease payments, augmented by “the difference between the fair market value of the asset and the guaranteed amount” (CBRE, 2016, p. 2); but the lessee is not required to recognize the full amount of the guarantee. Hence, financial engineers stress that “[t]he objective of structuring under the proposed rules will be to get the lowest present value/capitalized amount for the lessee” (Bosco, 2011, p. 2). In this way, a huge portion of a corporation’s actual outstanding lease obligation remains off balance sheet. Synthetic lease structures seem best suited to achieve this goal, as the high value of leased asset in these structures offers the lowest percentage of recognition (ELFA, 2018, p. 1).

Synthetic leases of the third generation (non-leveraged) strongly rely on the usage of residual guarantees. The residual risk mainly resides with the lessee, which allows the lessor to reduce the lease payments. Financial engineers exploit that the lessee only has to recognize the “amount, probable of being owed under residual value guarantees” (ASC 842–186) and accordingly design the lease contract in such a way that “the probable payment under the guarantee should be zero as the residual guarantee is generally structured with the strike price set at the expected future value—it is not ‘in the money’” (ELFA, 2018, p. 1). Additionally, there is no asset amortization (ELFA, 2017, p. 70) and all off-balance sheet non-lease or service components are excluded from the lease payments in order to minimize the PV. Furthermore, renewals only have to be recognized by the lessee if it is “reasonably certain” that the option is exercised, creating a regulatory loophole (see Biondi et al., 2011, pp. 864–865). Even though renewals get scrutiny by the lessee’s auditor, the big four accounting firms stated that (for real estate leases) a renewal is not automatically treated as compelling, if the base lease term exceeds five years or more. As a consequence, synthetic lease structures of the third generation are typically designed with a lease term between five to seven years in order to avoid any compelling renewal rents (Sebik, 2019). Hence, the new ASC 842 will only partially bring synthetic leases onto the lessee’s balance sheet, as shown in Table 1 below.

Table 1:

Exemplified Computation of Capitalization for Synthetic Leases (6% Discount Rate).

Lease type	Duration	Payments (monthly)	Value guarantee	Capitalized value
Low-value lease	3 Years	2.73%	–	≈91%
Synthetic lease	5 Years	0.5%	85%	≈26%

1. Low-value assets (e.g., Computer lease) have to be nearly completely consolidated under the new rules, while synthetic leases allow customers to keep a huge part of the lease obligation off balance sheet (ELFA, 2018: 2).

2. Source: Author’s table, based on ELFA (2018).

At the same time, tax laws remained largely unchanged over this period, and thus, synthetic leases still qualify as an on-balance sheet conditional sale for federal tax purposes. Figure 4 below depicts the basic mechanics of the revised synthetic lease structures:

Figure 4:

The Basic Structure of Synthetic Leasing (Third Generation).

The further shift of the residual risk from the lessor to the lessee might limit the range of potential customers to investment-grade corporations that are able to deal with the significant credit risk (is the lessee able to serve the residual guarantee, if needed?) (Borders et al., 2018, p. 3; ELFA, 2018, p. 1). However, due to the new accounting standards, providers of synthetic leases acknowledge an increasing demand for this sophisticated instrument (e.g., CBRE, 2016). This trend seems to be further fueled by the Tax Cuts and Jobs Act of 2017, which allows additional tax benefits for certain qualified tangible assets, other than real estate (Borders et al., 2018, p. 3; Kessler, 2019).

All in all, despite several attempts by accounting standard setters and regulators to close existing loopholes and to shift towards a more principles-based accounting approach, the combination of accounting standards and tax laws did not prevent a the perpetual evolution evolving survival perpetual evolution of synthetic leases over the last 30 years. The level of uncertainty for lessees remained manageable and the benefits of synthetic leases exceeded the regulatory costs for many users.

5 Discussion and Conclusion

This paper aimed to examine the interplay between accounting standards and tax laws in the context of regulatory arbitrage. For that purpose, the paper analyzed the perpetual “cycle of transactions innovation by preparers, followed by new or amended standards by standard setters, followed by further transactions innovations by preparers” (Dye et al., 2015, p. 272) for the case of synthetic leases in the USA. By examining a financial scheme through which managers seek “to get the best of both worlds” (Hanlon & Heitzman, 2010, p 133), this paper demonstrates how changing accounting standards enable financial engineers to exploit loopholes in the tax law and vice-versa. The article has shown that the gaming of rules may involve more than one regulatory frame and sheds light on the tense relationship between rules- and principles-based accounting standards and tax laws. Precise accounting standards and tax laws reduce uncertainties for financial engineers, as well as the regulatory burdens for preparers, while a two-book system allows managers to exploit the independent treatment of tax laws and accounting standards.

The emergence of synthetic leases in the early 1990s as part of the trend towards “creative accounting” (Nesvold, 1998) demonstrates that the introduction of rules-based accounting standards and tax laws reduced the level of uncertainty for financial engineering, shifting the regulatory cost-benefit analysis in favor of sophisticated off-balance sheet structures. The evolving structures of synthetic leases over the last 30 years shows the complementary interplay of increasingly principles-based accounting standards and rules-oriented tax laws. While research in the debate over principles- and rules-based regulation primarily either focuses on accounting standards (Donegan & Sunder, 1989; Maines et al., 2003b; Nelson, 2003) or tax laws (Avi-Yonah & Pichhadze, 2017; Kaplow, 1992; Weisbach, 1999), the case of synthetic leases shows that both regulatory frames are reciprocally linked (Biondi, 2019; Sunder, 2011), as changing tax laws motivated the financial engineering of accounting standards or the other way around. The revision of accounting standards in the aftermath of the Enron disaster only introduced new, precise bright lines, while the rules-based tax laws remained unchanged, enabling financial engineers to re-structure synthetic leases. Specifically, the recent revival of synthetic leases – in response to accounting standards that were expected to end the off-balance sheet treatment of leases – shows that tax laws and accounting standards have to be analyzed as two sides of the same coin. Figure 5 below depicts the link between precise accounting standards and tax laws.

Figure 5:

The Political Economy of Synthetic Leases.

In this vein, the paper also contributes to the discussion about book-tax conformity. Research in this area provides empirical evidence for the increasing gap between reported and tax income in the USA (Desai, 2005; Hanlon & Heitzman, 2010; Mills & Newberry, 2005); however, these studies say little about the relationship between off-balance sheet financing and tax optimization. The minimization of the lease exposure by means of synthetic leases strongly increases the residual risk for the lessee, thus imposing additional administrative burdens. While similar structuring efforts are also possible for other lease types, primarily synthetic leases seem to outweigh the high structuring costs, as they additionally offer tax benefits. Hence, tax laws shift the trade-off between costs and benefits of accounting-motivated financial engineering in favor of the latter. The evasion of ASC 842 by means of synthetic leases demonstrates that principles-based accounting standards, combined with a continuous approach (Donegan & Sunder, 1989; Dye et al., 2015), may not sufficiently raise the regulatory costs of regulatory arbitrage. Rather, the issue of boundary drawing is increasingly shifted to the auditors and their subjective assessments (Kunkel, 2020) while recent tax laws even facilitated the growth of off-balance sheet instruments.

This situation raises the question regarding whether a re-examination of the accounting consensus (Sunder, 2009) should also include the role of tax laws. Even in case of globally uniform, principles-based accounting standards, tax laws may still cause a diverging accounting treatment of similar transactions. However, this potential divergence can also be used as a means to limit regulatory arbitrage by applying selective book-tax conformity (Luppino, 2003; Raby & Richter, 1975) on certain transactions, mitigating unintended consequences of international accounting convergence.

A coordinative institutional setup is required to tackle regulatory arbitrage. A possible way forward might be to allow legislature, in close co-operation with national market regulators, the accounting standard setter and the tax authority, to impose book-tax conformity on certain off-balance sheet instruments. This close coordination may enable regulatory institutions to assess financial innovations from a holistic perspective and to choose between various regulatory means that are suited best to achieve regulatory goals. For example, instead of revising accounting standards, regulatory agencies may agree to impose additional taxes on certain structures. Moreover, such a coordinative approach can be fruitfully combined with an “embedded market regulator” (Thiemann & Lepoutre, 2017) that uses multi-polar supervisory dialogs between regulatory institutions and market actors in order to provide regulators with all relevant information (Thiemann & Troeger, 2020). The proposed regulatory approach may also include a retroactive application of tax law on financial innovations (Langenbucher, 2020) if these instruments have been identified by regulatory institutions for having no additional value in use – besides the evasion of regulatory requirements. Figure 6 below illustrates the proposed revision of the regulatory architecture:

Figure 6:

A Coordinative Approach in Accounting and Tax Regulation.

In a coordinative approach, accounting standard setter, securities regulators and tax authorities jointly decide which regulatory tool may suits best to achieve a desired accounting outcome. Besides further interpretations and the revision of accounting standards or tax laws, regulatory institutions might also demand legislation to impose selective book-tax conformity on certain transactions. This might be combined with a retroactive taxation on transactions which involve regulatory arbitrage.

In this way, the level of uncertainty increases for preparers, limiting regulatory arbitrage by making the structuring of certain transactions too costly, without having the negative effects of a universal book-tax conformity.