Safe Debt, Institutional Credibility and Financial Fragility: A Historical Governance Perspective on Public and Private Safe Assets

Safe assets occupy an unusual position in finance. They appear mundane in normal times because they circulate without elaborate inquiry, yet their disappearance during stress could define the boundary between ordinary volatility and systemic crisis. A safe asset is therefore best understood as a claim that can be transferred, pledged, or held at or near face value without requiring costly investigation into its underlying quality. This property is not equivalent to the absence of all risk. It is a social, legal, and institutional achievement that allows economic agents to economize on information production while transacting, saving, settling obligations, or posting collateral (G​o​r​t​o​n​,​ ​2​0​1​7; Holmstrom, 2015).

The historical record indicates that safety is never permanent. A medium regarded as unquestioned in one institutional environment may become suspect once its backing, legal status, or political guarantor is doubted. Precious-metal coins, bills of exchange, banknotes, deposits, sovereign bonds, asset-backed securities, and repurchase agreements have each performed safe-asset functions in particular contexts. None has done so simply because of its physical form or contractual label. Their acceptance has depended on minting standards, credible enforcement, fiscal capacity, convertibility, collateral quality, supervisory control, and collective belief in the continuity of these arrangements (N​o​r​t​h​ ​&​a​m​p​;​ ​W​e​i​n​g​a​s​t​,​ ​1​9​8​9; S​a​n​t​a​r​o​s​a​,​ ​2​0​1​5; Ugolini, 2017).

This article develops this argument within a governance and risk management framework. It examines safe assets as a form of financial infrastructure whose reliability was produced through public authority and the design of private contracts. The research was motivated by three observations. First, the earliest forms of safe assets did not solve verification problems; they merely made those problems more visible. Coins had metallic content, but users still faced uncertainty about weight, fineness, clipping, debasement, and counterfeiting (C​a​r​o​t​h​e​r​s​,​ ​1​9​3​0; M​u​n​r​o​,​ ​1​9​8​8; S​a​r​g​e​n​t​ ​&​a​m​p​;​ ​V​e​l​d​e​,​ ​1​9​9​9). Second, paper instruments increased transactional efficiency by substituting legal and reputational assurances for physical metal, but they introduced new vulnerabilities related to short maturity and confidence-sensitive redemption (Quinn & Roberds, 2015; Rogers, 1995). Third, modern financial systems have repeatedly relied on private instruments designed to mimic public safety, especially when public safe assets were scarce or when regulated deposits lost their dominance. The 2007–2008 crisis revealed the instability of such privately produced safe debt when collateral values and repo funding became information-sensitive (G​o​r​t​o​n​ ​&​a​m​p​;​ ​M​e​t​r​i​c​k​,​ ​2​0​1​2).

The study addresses three research questions. First, through which institutional mechanisms have different societies made financial claims acceptable at nominal value? Second, how has the composition of public and private safe debt shaped financial fragility across historical regimes? Third, what lessons can be drawn for corporate governance, insurance, banking supervision, and risk management? These questions are important because safe assets link board-level risk oversight, prudential regulation, liquidity insurance, collateral governance, and public policy.

The article makes three main contributions. It first repositioned safe-asset production as an institutional process rather than a technical classification of instruments. It then linked historical evidence on coins, bills, banknotes, sovereign debt, and securitization to the modern concepts of information insensitivity, collateral, pledgeability, and convenience yield. Finally, it translated these insights into risk management implications for intermediaries and regulators. The result was not an empirical test of a single episode; it was a structured historical synthesis designed to clarify the recurrent mechanisms through which confidence was created, expanded, and sometimes lost.

The remainder of the paper is organized as follows. Section 2 reviews the conceptual literature on safe assets and institutional credibility, while Section 3 explains the qualitative historical method. Section 4 traces the evolution of safe assets from metallic money to paper credit instruments and sovereign claims. Section 5 analyses the mechanisms through which debt acquires safe-asset properties. Section 6 discusses the public–private composition of safe debt and its relationship with financial fragility. Section 7 draws governance and risk management implications. Section 8 discusses safe assets as financial infrastructure, and Section 9 concludes the article.

The term “safety” can be misleading because, in ordinary language, a safe asset sounds like an instrument whose payoff is guaranteed. In financial history, however, safety is more closely related to acceptability. A claim performs a safe-asset function when market participants can use it without investing significant resources in discovering whether its issuer, collateral, or legal enforceability is impaired. G​o​r​t​o​n​ ​(​2​0​1​7​) described a safe asset as a claim accepted at face value without prolonged analysis, while Holmstrom (2015) emphasized the no-questions-asked character of money-like instruments. D​a​n​g​ ​e​t​ ​a​l​.​ ​(​2​0​1​7​) provided a related explanation: bank liabilities become useful when information about underlying assets is deliberately kept from being socially produced in ways that would destroy par exchange.

This understanding places safe assets between monetary theory and institutional economics. In monetary theory, safe assets provide liquidity, settlement capacity, collateral services, and stores of value. In institutional economics, those same functions depend on enforceable rules, credible public promises, and governance arrangements that limit opportunistic behavior (N​o​r​t​h​ ​&​a​m​p​;​ ​W​e​i​n​g​a​s​t​,​ ​1​9​8​9; S​t​a​s​a​v​a​g​e​,​ ​2​0​0​2). A coin, a note, or a Treasury bill is not simply a thing; it is a claim embedded in a system of authority, recognition, and enforcement.

The literature distinguished several related but non-identical attributes. Liquidity refers to the ease with which an asset can be converted into purchasing power with little impact on price. Moneyness refers to the capacity of a claim to circulate or settle obligations. Pledgeability refers to its usefulness as collateral. Safety, in a stricter sense, refers to the high probability that an instrument will retain its value. The convenience yield is the nonpecuniary return investors accept from holding instruments that provide these services. K​r​i​s​h​n​a​m​u​r​t​h​y​ ​&​a​m​p​;​ ​V​i​s​s​i​n​g​-​J​o​r​g​e​n​s​e​n​ ​(​2​0​1​2​) revealed that U.S. Treasuries had historically carried a yield advantage relative to comparable assets because investors valued their liquidity and safety services.

To avoid conceptual overlap, this article used safe assets as the broadest term for claims that can be held, pledged, or transferred at or near face value without costly verification. Safe debt refers to the debt-based subset of such claims, including sovereign bonds, deposits, bills, repos, and highly senior securitized liabilities. Money-like liabilities denote short-term claims used for settlement, collateral, or liquidity management, but they are not necessarily safe unless institutional or collateral backing preserves par acceptance. Information-insensitive claims refer to instruments whose holders have little incentive to produce private information before accepting them. These terms overlap because many safe debts are money-like and information-insensitive in normal times; however, the overlap is conditional on the institutional mechanisms that sustain confidence under stress.

Institutional credibility is the bridge between these characteristics. Public safe assets usually rely on taxation power, monetary authority, legal continuity, and political commitment. Private safe assets rely on collateral, seniority, diversification, short maturity, convertibility, monitoring, and sometimes explicit or implicit public support. The difference is not absolute. Sovereigns may default, inflate, restructure, or repudiate. Private claims may be robust when legal protections and collateral regimes are strong. Yet the mechanisms differ: Public safe assets are backed by the fiscal and legal capacity of the state, whereas private safe assets are engineered through contractual structures and balance-sheet design.

A central problem is that the private production of money-like debt could become excessive. When financial intermediaries issue short-term liabilities against longer-term or opaque assets, the system may appear liquid until confidence changes. S​t​e​i​n​ ​(​2​0​1​2​) modeled this as an externality: Unregulated creation of private money can leave the financial system too exposed to costly runs. The problem is not only insolvency. A system can become fragile because the assets backing short-term claims become difficult to value precisely when valuation is most needed.

The scarcity of safe assets adds a macro-financial dimension. C​a​b​a​l​l​e​r​o​ ​(​2​0​0​6​) and C​a​b​a​l​l​e​r​o​ ​e​t​ ​a​l​.​ ​(​2​0​1​7​) connected the shortage of credible stores of value to low interest rates, global imbalances, and safety-trap dynamics. G​o​r​t​o​n​ ​&​ ​O​r​d​o​ñ​e​z​ ​(​2​0​2​2​) similarly framed safe assets as publicly and privately supplied claims demanded for saving and collateral purposes. When public supply is inadequate or unevenly distributed, private actors create substitutes. This substitution may satisfy liquidity demand in tranquil periods, but it may also increase fragility if the private substitutes lose their information-insensitive status in stress.

Recent evidence further clarifies how financial intermediaries produce privately issued safe assets. Kacperczyk et al. (2021) show that private debt instruments can acquire safe-asset characteristics when their issuers possess strong balance sheets and when institutional arrangements support liquidity and repayment at par. Their findings indicate that privately produced safety is neither automatic nor determined solely by contractual seniority; it also depends on issuer quality, market confidence, and the credibility of public and private backstops. This evidence complements the framework of G​o​r​t​o​n​ ​&​ ​O​r​d​o​ñ​e​z​ ​(​2​0​2​2​), in which public and private safe assets jointly respond to the demand for stores of value and collateral. A shortage of credible public claims may therefore stimulate private safe-asset production, while simultaneously increasing the financial system’s exposure to confidence-sensitive liabilities.

For governance and risk management, these arguments implied that safe assets should not be assessed only by credit ratings, capital charges, or contractual seniority. Their safety depends on whether the institutional environment could sustain confidence when incentives to produce private information rise. This approach converts the safe-asset problem into a question of board oversight, supervisory design, liquidity insurance, collateral valuation, disclosure, and capacity of crisis management.

This article adopted a qualitative historical-analytical method, not with the aim of estimating a causal coefficient or testing a time-series relationship. Instead, the study identified recurrent mechanisms in the historical development of safe assets and interpreted them through the concepts of information insensitivity, collateral, convenience yield, public credibility, and financial fragility. This approach was appropriate because the core object of analysis and the perceived safety of financial claims were historically contingent and institutionally embedded.

The source base consisted of economic history studies on coinage, bills of exchange, early banking, sovereign debt, and financial crises. It also involved theoretical and empirical work on safe assets, bank runs, and collateral as well as policy literature on liquidity risk and non-bank financial intermediation. The analysis gave particular attention to instruments and arrangements that appeared repeatedly in the safe-asset literature. The topics included metallic coinage, public banks, bills of exchange, goldsmith notes, demand deposits, sovereign debt, securitized claims, repos, and regulatory support such as deposit insurance and entry restrictions.

The synthesis proceeded in four steps. First, historical instruments were examined according to the source of trust on which their circulation depended. Second, the main frictions attached to each instrument were identified, including verification costs, adverse selection, collateral opacity, convertibility risk, and political default risk. Third, the article compared public and private mechanisms for producing safe debt. Fourth, the historical findings were translated into governance and risk management implications relevant to banks, insurers, institutional investors, supervisors, and fiscal authorities.

The study was deliberately selective as it focused on mechanisms rather than offering a complete global monetary history. China, the Mediterranean, Western Europe, the Ottoman Empire, Britain, and the United States were included as reference points insofar as they illuminated institutional patterns. The article did not reproduce historical narratives for their own sake; it used them to show why the safety of financial claims must be understood as a governance outcome.

The historical episodes were selected according to four criteria. First, each episode had to represent a distinct mechanism through which nominal acceptance was created or lost, such as metallic certification, commercial law, sovereign fiscal commitment, deposit insurance, or collateralized wholesale funding. Second, each episode had to be influential in the evolution of safe-asset production and frequently discussed in the economic-history or safe-asset literature. Third, the case had to provide identifiable institutional evidence on credibility, enforcement, convertibility, collateral governance, or public backing. Fourth, it had to be theoretically relevant for comparing public and private mechanisms of safety. The cases were therefore illustrative and mechanism-oriented rather than exhaustive; they were used to discipline the historical synthesis and to reduce subjectivity in case selection.

The first major safe-asset regime was based on precious metals. Coins provided recognizable units, portability, divisibility, and a connection to metal content. This made them valuable in exchange and debt settlement, especially before credible paper money or transferable credit instruments became widespread (H​o​p​p​i​t​,​ ​1​9​8​6; Vilar, 2025). Nevertheless, the existence of metal did not eliminate information problems. It merely shifted them to questions of purity, weight, denomination, and official certification.

Ancient coinage is often associated with Lydia, although the historical geography of early coinage is complex. Howgego (1995) emphasized the importance of Lydian electrum coins in early archaeological contexts, while D​a​v​i​e​s​ ​(​2​0​0​2​), Peng (1994), T​u​l​l​o​c​k​ ​(​1​9​5​7​), and Yang (1952) underlined the independent significance of Chinese monetary innovations, including paper money and state control over paper currency. What mattered for the present argument was not the exact place of invention, but the institutional lesson: Coinage became useful when users could recognize a standard that reduced the cost of valuation.

That standard was imperfect. Port cities often hosted confusing mixtures of foreign units. Small-change shortages created transactional frictions. More importantly, coins could be clipped, sweated, debased, counterfeited, cried up, or cried down (C​a​r​o​t​h​e​r​s​,​ ​1​9​3​0; K​i​n​d​l​e​b​e​r​g​e​r​,​ ​1​9​9​1; Mate, 1972; R​e​d​i​s​h​,​ ​1​9​9​0). Debasement was particularly damaging because it broke the link between nominal value and expected metallic content (M​u​n​r​o​,​ ​1​9​8​8; R​o​l​n​i​c​k​ ​e​t​ ​a​l​.​,​ ​1​9​9​6). The result was a monetary environment in which users could not always accept coins at face value. They had to weigh, assay, inspect, or rely on specialists.

The cost of such verification is central to the history of safe assets. A coin may be physically durable and intrinsically valuable, but it is not fully safe if every transaction requires costly quality assessment. Assaying requires expertise and equipment, while weighing requires trusted scales. In this respect, coinage already displays the basic safe-asset problem: Economic agents want instruments that would circulate without repeated information production. Mint marks, legal tender rules, recoinage, and public certification attempt to solve this problem by moving trust from metal alone to institutional authority (Price, 1869).

The difficulties of coinage contributed to the creation of public banks and other money-like institutions. These institutions try to replace heterogeneous metallic circulation with more standardized claims, although they are not always state banks in the modern sense (Roberds & Velde, 2016a; Roberds & Velde, 2016b; Ugolini, 2017). Their historical role is to centralize verification and convert uncertain metal into more reliable book claims or circulating liabilities. The shift from material inspection to institutional certification had already begun.

Recent research on the Bank of Amsterdam provides further evidence for this transition from material value to institutional trust. B​o​l​t​ ​e​t​ ​a​l​.​ ​(​2​0​2​4​) show that the bank’s ledger money achieved wide acceptance because a stable unit of account, standardized settlement procedures, public sponsorship, and disciplined governance reduced the need to inspect the metallic content of individual coins. In this sense, the bank operated as a public payment infrastructure that converted heterogeneous and potentially debased metallic media into standardized book money. Its historical experience also demonstrates the limits of institutionally created money: the credibility of fiat-like claims remained conditional on the quality of the assets, governance arrangements, and public commitments supporting the issuing institution.

The second transition was from metal-based settlement to transferable credit instruments. Bills of exchange, goldsmith receipts, banknotes, checks, and deposits emerged because trade required instruments that could move value across distance and time more efficiently than bags of coin. These instruments were paper claims, but their credibility depended on legal enforceability, commercial reputation, endorsement practices, and convertibility into recognized money (d​e​ ​R​o​o​v​e​r​,​ ​1​9​5​3; Rogers, 1995; S​a​n​t​a​r​o​s​a​,​ ​2​0​1​5; Usher, 1914).

Bills of exchange illustrate how private law could create money-like circulation. A bill connects an issuer, a payer, and a beneficiary across locations, and it could pass through successive endorsers before maturity. Each endorsement adds a layer of liability and reputation. This joint liability makes the bill more acceptable, but it also limits its anonymity: Holders often had to evaluate the credit quality of endorsers before accepting it (G​e​v​a​,​ ​2​0​1​1; Munro, 2003; Rogers, 1995). Thus, bills are not pure no-questions-asked assets. They reduce metal-transfer costs but replace them with counterparty and legal-information costs.

By the eighteenth century, bills increasingly functioned as substitutes for money in commercial networks. This development expanded liquidity but also made private credit networks vulnerable to sudden reversals. The 1763 crisis in northern Europe, triggered by the failure of a major banking house and transmitted through short-term merchant-bank funding, demonstrated that a privately produced instrument could be liquid in normal times and fragile in crisis (Quinn & Roberds, 2015). The historical parallel with modern wholesale funding is strong: confidence in a debt chain depended on the continuing acceptability of its backing relationships.

Goldsmith notes and early banknotes added another trust mechanism. Their circulation depended less on the last endorser and more on the known identity and standing of the issuer. In London, goldsmiths held precious metal in safes and issued receipts that could circulate (Q​u​i​n​n​,​ ​1​9​9​7; Richards, 1929; Temin & Voth, 2006). Thornton (2017) distinguished the confidence placed in the endorser of a bill from the more general reputation supporting a banknote. This distinction remains analytically important. Some safe assets circulate because the identity of prior parties is credible; others circulate because the issuing institution itself is trusted.

Demand deposits and checks further transform safe assets by moving circulation from individual paper instruments to bank balance sheets. Yet the central question remains unresolved: How can holders of short-term claims be confident that conversion into money will be honored? Convertibility and short maturity serve as discipline devices, while collateral, reserves, unlimited liability, clearinghouse monitoring, and later deposit insurance serve as confidence technologies (C​a​l​o​m​i​r​i​s​ ​&​a​m​p​;​ ​K​a​h​n​,​ ​1​9​9​6; D​i​a​m​o​n​d​ ​&​a​m​p​;​ ​R​a​j​a​n​,​ ​2​0​0​0; T​i​m​b​e​r​l​a​k​e​,​ ​1​9​8​4). Each mechanism reduces uncertainty, but none removes the possibility of a run when confidence collapses.

The public production of safe assets developed rather slowly. Early sovereign borrowing often financed war, and repayment depended on coercion, taxation, confiscation, monetary debasement, or future military success. Sovereigns frequently defaulted or restructured. Consequently, public debt did not become safe merely because the borrower was a state. It became safer when institutions constrained rulers, stabilized revenue, and made repayment credible (N​o​r​t​h​ ​&​a​m​p​;​ ​W​e​i​n​g​a​s​t​,​ ​1​9​8​9; S​t​a​s​a​v​a​g​e​,​ ​2​0​0​2).

England is central in the literature because the post-Glorious Revolution settlement strengthened parliamentary control over public finance and improved the credibility of debt commitments. N​o​r​t​h​ ​&​a​m​p​;​ ​W​e​i​n​g​a​s​t​ ​(​1​9​8​9​) famously interpreted this as a constitutional transformation that limited arbitrary default. Later studies nuanced the timing and channels of this credibility effect. S​u​s​s​m​a​n​ ​&​a​m​p​;​ ​Y​a​f​e​h​ ​(​2​0​0​6​) argued that borrowing costs did not adjust immediately, while Stasavage highlighted the political distribution of creditors and the role of elite support (S​t​a​s​a​v​a​g​e​,​ ​2​0​0​2; Stasavage, 2011). The common point is that sovereign debt becomes safe through institutional arrangements, not through sovereignty alone.

Private intermediaries also mattered. In the nineteenth century, underwriters and merchant banks lent their reputations to sovereign borrowers to help investors interpret political risk and fiscal capacity (F​l​a​n​d​r​e​a​u​ ​&​a​m​p​;​ ​F​l​o​r​e​s​,​ ​2​0​0​9; F​l​a​n​d​r​e​a​u​ ​&​a​m​p​;​ ​Z​u​m​e​r​,​ ​2​0​0​9). Thus, the public-private boundary in safe-asset production has always been porous. Public debt requires private certification, while private debt requires public law and lender-of-last-resort expectations.

The Ottoman experience reinforces the institutional argument. Ottoman borrowing in the eighteenth and nineteenth centuries shows that market acceptance of sovereign debt depended on fiscal capacity, administrative credibility, and external confidence as much as on the formal debt instrument itself. Episodes of borrowing pressure, restructuring, and default illustrate that sovereign claims become credible only when investors believe that the state has durable repayment capacity and enforceable fiscal discipline (A​r​s​l​a​n​,​ ​2​0​1​5; Kopar & Yolun, 2012). This case is analytically valuable because it connects public debt safety with fiscal governance, institutional reputation, and the political economy of commitment.

Table 1 summarizes the historical progression. The same pattern recurs across different instruments: A claim becomes safe when a mechanism lowers the costs of verification and stabilizes expectations. Yet each mechanism creates its own fragility.

The historical discussion shows that the safety of debt is constructed through recurring mechanisms. These mechanisms do not eliminate risk; they change the distribution of information, incentives, and losses. Four mechanisms are especially significant: information insensitivity, collateral and seniority, convertibility or short maturity, and institutional insurance.

Information insensitivity means that agents have little reason to investigate the underlying value of a claim before accepting it. This is not ignorance in a simple sense; it is an equilibrium property. If everyone expects a claim to be accepted at face value, and if producing private information is costly or unprofitable, the claim circulates smoothly. If the expected benefit of investigation increases, the same instrument can become information-sensitive and cease to function as a safe asset (D​a​n​g​ ​e​t​ ​a​l​.​,​ ​2​0​1​7; G​o​r​t​o​n​,​ ​2​0​1​7).

Coins become information-sensitive when their metallic content is uncertain. Meanwhile, bills become information-sensitive when the reliability of endorsers is questioned. Bank deposits become information-sensitive when depositors fear that assets backing bank liabilities are insufficient. Repo claims become information-sensitive when the value of securitized collateral becomes uncertain. The forms differ, but the transition is similar: Safety is lost when agents begin to ask questions that the system is designed to make it unnecessary.

This point also explains why transparency is not always a simple remedy. In ordinary corporate governance, greater disclosure often improves accountability. In the production of safe debt, however, poorly designed disclosure can undermine par circulation if it encourages sudden differentiation among claims that have been treated as homogeneous. The regulatory challenge is therefore not to hide risk, but to decide which information should be disclosed to whom, when, and in what form. Supervisors need enough information to control systemic risk, while money-like instruments require stable conventions that prevent constant adverse-selection bargaining.

Collateral is the most visible mechanism to produce private safe debt. A debt claim becomes safer when it is backed by assets that are themselves liquid, durable, and legally enforceable. Land, specie, government bonds, loan portfolios, mortgages, asset-backed securities, and other collateral pools have all served this function in different periods. B​a​n​k​ ​f​o​r​ ​I​n​t​e​r​n​a​t​i​o​n​a​l​ ​S​e​t​t​l​e​m​e​n​t​s​ ​(​2​0​0​1​) emphasized that collateral in wholesale markets was not merely a credit-risk mitigant; it also shaped market liquidity, risk management, and market dynamics.

Collateral, however, is not self-validating. Its value depends on legal title, custody, valuation conventions, margining rules, market depth, and the correlation of collateral prices with the borrower’s own condition. During stress, collateral that seems safe may become difficult to price or sell. Haircuts rise, margins increase, and leverage must be unwound. The F​i​n​a​n​c​i​a​l​ ​S​t​a​b​i​l​i​t​y​ ​B​o​a​r​d​ ​(​2​0​2​3​) noted that leverage in non-bank financial intermediation could generate large liquidity demands through collateral and margin calls, forcing asset sales that amplify volatility. This is a modern expression of an older pattern: Collateral provides safety until confidence in collateral valuation becomes the channel of panic.

Seniority complements collateral. Because debt claims rank ahead of equity and state a fixed payment promise, their payoff is less exposed to ordinary information about upside potential. Short-term senior claims can therefore appear stable. Yet priority concentrates losses on junior claimants and may give senior holders incentives to exit before others. A run is not irrational simply because a claim is senior; it may be rational precisely because senior claimants fear that collateral coverage will deteriorate if they wait (D​i​a​m​o​n​d​ ​&​a​m​p​;​ ​D​y​b​v​i​g​,​ ​1​9​8​3; G​o​r​t​o​n​ ​&​a​m​p​;​ ​M​e​t​r​i​c​k​,​ ​2​0​1​2).

Convertibility and short maturity have historically been used as monitoring devices. If a note, deposit, or repo can be redeemed quickly, the issuer is pressured to maintain liquid backing. C​a​l​o​m​i​r​i​s​ ​&​a​m​p​;​ ​K​a​h​n​ ​(​1​9​9​6​) and D​i​a​m​o​n​d​ ​&​a​m​p​;​ ​R​a​j​a​n​ ​(​2​0​0​0​) explained why demandable debt could discipline intermediaries: The threat of withdrawal constrained managerial misuse of assets. Early goldsmith notes, banknotes, and demand deposits relied heavily on the belief that holders could convert claims into specie or money on demand.

The same feature generates fragility. If a number of holders demand conversion simultaneously, the issuer may need to sell assets to falling markets. When asset sales depress prices, mark-to-market losses can spread to other institutions. The 2007–2008 experience with repo and securitized assets demonstrated how short-term secured funding could transmit doubts about collateral to system-wide deleveraging (G​o​r​t​o​n​ ​&​a​m​p​;​ ​M​e​t​r​i​c​k​,​ ​2​0​1​2). The Basel Committee’s post-2023 analysis similarly stressed the importance of liquidity risk management, governance, and the speed at which confidence shocks could generate deposit outflows in modern banking (B​a​s​e​l​ ​C​o​m​m​i​t​t​e​e​ ​o​n​ ​B​a​n​k​i​n​g​ ​S​u​p​e​r​v​i​s​i​o​n​,​ ​2​0​2​3; B​a​s​e​l​ ​C​o​m​m​i​t​t​e​e​ ​o​n​ ​B​a​n​k​i​n​g​ ​S​u​p​e​r​v​i​s​i​o​n​,​ ​2​0​2​4).

Thus, convertibility is both a source of discipline and a source of instability. It supports trust when redemption requests are dispersed; it can destroy trust when redemption becomes collective. This dual role is why private safe debt usually requires regulation, lender-of-last-resort capacity, deposit insurance, or other public backstops.

Safe assets offer services that are not captured entirely by their cash flows. Investors hold them because they provide liquidity, collateral services, regulatory value, transaction services, and crisis-time protection. The yield concession that investors accept in exchange for these services is the convenience yield. K​r​i​s​h​n​a​m​u​r​t​h​y​ ​&​a​m​p​;​ ​V​i​s​s​i​n​g​-​J​o​r​g​e​n​s​e​n​ ​(​2​0​1​2​) estimated that Treasury securities received a substantial yield benefit from their safety and liquidity characteristics. This benefit was not a puzzle once safety was understood as a service.

The convenience yield also guides private safe-asset production. When the demand for safety rises, intermediaries have incentives to manufacture instruments that resemble public safe assets. Securitization, tranching, repo collateralization, and special purpose vehicles can all be interpreted as attempts to create claims with low information sensitivity and high pledgeability (G​o​r​t​o​n​ ​&​a​m​p​;​ ​M​e​t​r​i​c​k​,​ ​2​0​1​3; Gorton & Souleles, 2007). In tranquil periods, this can improve credit supply. In fragile periods, however, the same structures can concentrate opacity and leverage.

For regulators, convenience yield has an important implication. The private sector will not stop producing safe-asset substitutes simply because they create systemic externalities. If safe assets command a price premium, financial innovation will search for ways to capture it. Governance must therefore focus on the conditions under which safe-debt creation is socially useful and the point at which it becomes destabilizing.

The relative supply of public and private safe debt is one of the most crucial determinants of financial fragility. Public debt that is credibly backed by fiscal capacity can provide benchmark collateral and reduce the need for fragile private substitutes. Private safe debt can improve liquidity and credit intermediation, but it is more dependent on collateral valuation, rollover confidence, and public support. The challenge is not that private safe debt is always problematic. The private supply of safety tends to be procyclical: It expands when confidence is high and contracts precisely when safety is most valuable.

The distinction between the production of public and private safe assets can be summarized by the source of credibility, the dominant risk channel, the governance mechanism, and the typical failure mode. Public safety rests primarily on fiscal and legal credibility, whereas private safety is engineered through contractual and collateral structures that can become fragile when confidence changes.

Table 2 summarizes these differences in terms of the source of credibility, the dominant risk channel, the governance mechanism, and the typical failure mode.

As shown in Table 2, public safe debt derives its credibility primarily from fiscal capacity, legal continuity, monetary authority, and political commitment, whereas private safe debt relies more heavily on collateral quality, contractual seniority, issuer reputation, diversification, and convertibility. Consequently, public safe debt is mainly exposed to fiscal stress, inflation, default, restructuring, and credibility loss, while private safe debt is more vulnerable to rollover risk, collateral revaluation, margin calls, opacity, and runs. The comparison therefore demonstrates that both forms of safe debt require credible governance arrangements, although private safety depends more directly on continuous market confidence.

Government debt becomes safe when investors believe that the state can and will honor the claim. This requires more than nominal sovereignty. It requires revenue capacity, legal continuity, policy credibility, political constraints on opportunism, and access to monetary arrangements that do not destroy real value. The classic case of England showed how constitutional and fiscal institutions could transform the risk profile of public borrowing (N​o​r​t​h​ ​&​ ​W​e​i​n​g​a​s​t​,​ ​1​9​8​9; S​t​a​s​a​v​a​g​e​,​ ​2​0​0​2; S​u​s​s​m​a​n​ ​&​ ​Y​a​f​e​h​,​ ​2​0​0​6).

Public safe assets support financial stability by serving as collateral, liquidity buffers, and benchmarks for pricing. They can reduce adverse selection because their backing is tied to the consolidated fiscal capacity of the state rather than the quality of a private asset pool. However, public safe-asset supply is not unlimited. Excessive debt, weak fiscal institutions, inflationary finance, or political instability could reduce safety. This creates a fiscal governance challenge: The public sector must supply enough credible safe assets to support financial intermediation without undermining the credibility on which their safety depends (C​a​b​a​l​l​e​r​o​ ​e​t​ ​a​l​.​,​ ​2​0​1​7; G​o​r​t​o​n​ ​&​ ​O​r​d​o​ñ​e​z​,​ ​2​0​2​2).

Recent market evidence shows that even public instruments normally treated as safe may temporarily lose some of their liquidity and collateral services during severe stress. H​e​ ​e​t​ ​a​l​.​ ​(​2​0​2​2​) document that U.S. Treasury securities exhibited “inconvenience yields” during the COVID-19 market turmoil, as selling pressure and constrained dealer balance sheets disrupted market functioning. H​a​d​d​a​d​ ​e​t​ ​a​l​.​ ​(​2​0​2​1​) similarly show that liquidity disturbances spread across debt markets when investors attempted to liquidate similar securities simultaneously and that central bank intervention contributed to the restoration of market pricing. These findings distinguish credit safety from market liquidity: a sovereign instrument may remain highly creditworthy while becoming difficult or costly to trade. Private safe-debt substitutes are even more vulnerable because their acceptability depends not only on issuer credibility but also on dealer capacity, collateral valuation, haircuts, and uninterrupted short-term funding.

Private safe debt expands when intermediaries transform risky or illiquid assets into claims that appear standardized, senior, and collateralizable. Modern securitization does this by pooling assets, creating tranches, using special purpose vehicles, and assigning priority structures. Asset-backed securities and mortgage-backed securities were designed to convert heterogeneous loan cash flows into marketable claims. When highly rated tranches were funded by repo and other short-term liabilities, the system created money-like instruments outside traditional deposit banking (G​o​r​t​o​n​ ​&​a​m​p​;​ ​M​e​t​r​i​c​k​,​ ​2​0​1​2; Gorton & Souleles, 2007).

This structure depends on the belief that senior claims are sufficiently protected and that collateral can be liquidated or financed without large discounts. Once doubts emerge, information sensitivity returns. Investors no longer treat apparently homogeneous claims as equivalent. Haircuts rise, funding becomes more difficult to roll over, and asset sales depress prices. The private production of safe debt therefore contains an internal tension: It works by suppressing information sensitivity, but its fragility appears when market participants have incentives to rediscover information all at once.

The development of shadow banking can be read through this lens. Shadow banking does not arise simply because regulation is absent. It also arises because the financial system demands money-like collateral and because regulated deposits are no longer the sole dominant private safe asset. The decline in charter value, competition from money market funds and high-yield markets, as well as innovations in securitized finance change the institutional structure of private liquidity creation (Gorton & Rosen, 1995; Keeley, 1990; Moreira & Savov, 2017).

A run is often described as a sudden withdrawal of funds. Fundamentally, it is the collective collapse of an arrangement that allows claims to circulate without verification. Historical runs on bills, banknotes, deposits, and repo differ in institutional form, but they share a mechanism: Holders stop treating the claim as safe and demand a more trusted settlement asset. This demand for conversion forces the issuer to liquidate, obtain public support, or suspend convertibility.

Runs are therefore governance failures as much as market events. They reflect inadequate liquidity planning, weak collateral governance, flawed incentive structures, or supervisory gaps. Deposit insurance could stabilize retail deposits by replacing private confidence with a public guarantee, but it may also create moral hazard if pricing and supervision are weak (C​a​l​o​m​i​r​i​s​,​ ​1​9​9​0; D​i​a​m​o​n​d​ ​&​a​m​p​;​ ​D​y​b​v​i​g​,​ ​1​9​8​3). Clearinghouses historically managed member risk and mutual support, but their effectiveness depended on credible monitoring and collective discipline (G​o​r​t​o​n​,​ ​1​9​8​5; T​i​m​b​e​r​l​a​k​e​,​ ​1​9​8​4). Modern lender-of-last-resort facilities similarly require rules that distinguish liquidity support from solvency subsidy.

The 2023 banking turmoil renewed attention to governance failures in liquidity risk management. The Basel Committee emphasized that rapid deposit outflows, uninsured funding concentrations, interest-rate risk, and weak management practices could interact with modern communication technologies (B​a​s​e​l​ ​C​o​m​m​i​t​t​e​e​ ​o​n​ ​B​a​n​k​i​n​g​ ​S​u​p​e​r​v​i​s​i​o​n​,​ ​2​0​2​3; B​a​s​e​l​ ​C​o​m​m​i​t​t​e​e​ ​o​n​ ​B​a​n​k​i​n​g​ ​S​u​p​e​r​v​i​s​i​o​n​,​ ​2​0​2​4). The lesson is consistent with historical evidence: The faster claims can be questioned and withdrawn, the more demanding the governance of safety becomes.

Recent empirical research also cautions against identifying banking fragility only with visible depositor panics. Using long-run cross-country evidence, B​a​r​o​n​ ​e​t​ ​a​l​.​ ​(​2​0​2​1​) show that banking crises may develop through declining bank equity values, credit contractions, and deteriorating balance-sheet conditions even when a conventional depositor panic is absent. Read together with the 2023 banking turmoil, this evidence suggests that boards and supervisors should monitor latent valuation losses, uninsured funding concentrations, interest-rate exposures, and access to contingent liquidity before withdrawals become observable. Effective governance should therefore combine market-based early-warning indicators with deposit-run scenarios, collateral stress tests, and credible contingency funding plans.

Safe assets matter directly for corporate governance, insurance, and risk management because they shape the solvency and liquidity of financial institutions. Boards and risk committees often treat highly rated or highly liquid instruments as low-risk items on the balance sheet. The historical perspective suggests that this treatment is incomplete. A safe asset must be evaluated not only by issuer credit quality but also by the institutional conditions that keep it acceptable during stress.

Corporate governance frameworks should require boards of banks, insurers, pension funds, and asset managers to distinguish between accounting safety, market liquidity, and systemic liquidity. An asset may be of high credit quality but illiquid under stress. A collateral pool may be diversified but exposed to correlated margin calls. A short-term funding structure may be profitable but unstable if counterparties can withdraw simultaneously. These issues belong in board risk appetite statements, internal stress testing, and contingency funding plans.

Owing to the paramount importance of collateral governance, institutions should monitor not only nominal collateral values but also haircut sensitivity, concentration, legal enforceability, wrong-way risk, rehypothecation chains, and market depth. The Financial Stability Board has stressed that leverage in non-bank financial intermediation could amplify stress through liquidity demands and asset sales (F​i​n​a​n​c​i​a​l​ ​S​t​a​b​i​l​i​t​y​ ​B​o​a​r​d​,​ ​2​0​2​3; F​i​n​a​n​c​i​a​l​ ​S​t​a​b​i​l​i​t​y​ ​B​o​a​r​d​,​ ​2​0​2​4). This means that risk managers must model collateral as a dynamic source of liquidity demand, not merely as static protection.

Insurance mechanisms are central to the production of safe assets. Deposit insurance transforms privately issued bank liabilities into claims supported by public credibility. Lender-of-last-resort facilities provide liquidity insurance to solvent institutions facing runs. Central bank swap lines and market-wide facilities could stabilize collateral markets. Yet each insurance mechanism creates moral hazard unless accompanied by supervision, capital requirements, liquidity standards, resolution regimes, and credible loss allocation.

The historical lesson is that insurance should be priced and governed according to the systemic services being provided. When a bank issues money-like liabilities, it produces a public service but may also impose public costs. Regulation should therefore align private incentives with the social objective of stable liquidity. Charter value, entry restrictions, capital buffers, liquidity coverage, net stable funding rules, and resolution planning are different methods for managing this alignment (M​a​r​c​u​s​,​ ​1​9​8​4; S​h​l​e​i​f​e​r​ ​&​a​m​p​;​ ​V​i​s​h​n​y​,​ ​2​0​1​0; S​t​e​i​n​,​ ​2​0​1​2).

Public debt management is usually treated as a fiscal or cost-minimization problem. The safe-asset perspective adds a financial-stability dimension. The maturity, liquidity, and credibility of government securities influence the supply of benchmark collateral available to the financial system. Greenwood et al. (2015) argued that the government had comparative advantages in supplying certain forms of safe debt. When public safe-asset supply is insufficient, private substitutes may proliferate. When credibility of public debt weakens, the entire collateral hierarchy may deteriorate.

Macroprudential policy should therefore coordinate fiscal authorities, central banks, and supervisors. Fiscal authorities influence the quantity and maturity structure of public safe assets. Central banks influence the composition of outstanding money-like claims through open market operations and lender-of-last-resort policies. Supervisors influence the amount of private safe debt through capital, liquidity, and collateral rules. Treating these policies separately risks producing unintended shifts from regulated to unregulated safe-debt creation.

Table 3 translates the historical findings into practical questions for boards, regulators, and risk professionals.

The main finding in this article is that safe assets are not natural objects but institutional infrastructure. Their production requires a chain of arrangements that make it unnecessary for every holder to inspect every asset at every moment. Minting standards, commercial law, sovereign fiscal commitment, clearinghouse rules, deposit insurance, collateral law, prudential regulation, and central bank support are all historically specific ways of producing the same outcome: par acceptance in the face of uncertainty.

This interpretation changes how financial fragility should be understood. Fragility does not arise only because institutions hold risky assets. It also arises because they issue claims whose value depends on not being questioned. A balance sheet may be fragile precisely because it appears safe in ordinary times. The more an instrument is used as money-like collateral, the greater the system-wide consequences when its safety is doubted.

The historical movement from coins to bills, from bills to bank liabilities, from bank liabilities to sovereign debt, and from sovereign debt to securitized private substitutes does not represent linear progress from primitive to modern finance. It represents successive attempts to reduce verification costs and expand liquidity. Each solution creates a new governance problem. Coins require certification and bills require legal transferability. Banknotes require convertibility and issuer reputation. Deposits require insurance and supervision. Sovereign debt requires fiscal commitment. Securitized safe debt requires collateral governance and systemic liquidity backstops.

This means that the balance between public and private safe debt is a policy variable, not a background condition. Public safe assets could anchor the financial system when credible, but their supply depends on fiscal capacity and political legitimacy. Private safe assets could fill collateral demand and support credit, but their safety is conditional and procyclical. A stable system therefore requires neither the elimination of private safe-debt production nor unlimited public debt issuance. It requires governance of the interface between the two.

Recent episodes demonstrate the contemporary relevance of this historical logic. During the 2007–2008 crisis, repo-financed securitized claims lost their money-like status once collateral valuation became contested. The 2023 banking turmoil similarly revealed how uninsured deposit concentrations, interest-rate losses, and rapid withdrawals could turn apparently stable liabilities into information-sensitive claims. Recent leverage and margin-call concerns in non-bank financial intermediation have reinforced the same point: Modern safe-asset stress begins when governance arrangements supporting par acceptance are questioned.

This study has reframed the safe-asset question from a historical governance perspective. It has argued that the safety of financial claims depends on institutional credibility, legal enforceability, collateral design, and public-private coordination rather than intrinsic value. The long history of coins, bills, banknotes, deposits, sovereign debt, and securitized claims shows that every safe-asset regime solved one information problem while creating another.

The current analysis yielded four conclusions. First, metallic value alone did not create fully safe assets because coins required costly verification and were vulnerable to debasement and manipulation. Second, paper credit instruments expanded liquidity by replacing physical metal with legal and reputational mechanisms, but their reliance on short maturity and convertibility made them run-prone. Third, sovereign debt became safe only where fiscal capacity and political commitment made repayment credible. Fourth, modern private safe-debt production could satisfy demand for collateral and liquidity, but it could also amplify financial fragility when public safe assets were scarce and private substitutes became information-sensitive.

For policy and corporate governance, the implication is direct. Safe-asset production should be managed as a systemic risk issue. Boards must monitor the liquidity and collateral assumptions behind apparently safe instruments. Supervisors must regulate private money creation where it produces public externalities. Fiscal and monetary authorities must recognize that public debt management and central bank operations influence the incentives for private safe-debt creation. Deposit insurance and liquidity backstops must be accompanied by prudential safeguards to prevent moral hazard.

The study’s qualitative design and selective historical scope impose several limitations. It did not estimate the quantitative effect of public safe-asset supply on private safe-debt creation, test country-level differences in fiscal credibility, or establish the relative causal importance of the mechanisms identified. Moreover, the analysis may be subject to regional bias because the cases were drawn primarily from contexts with comparatively well-developed economic-historical and institutional evidence, particularly Europe, the North Atlantic, the Ottoman Empire, China, and the United States. Accordingly, the study does not claim to represent all regional monetary trajectories in equal depth or to establish generalizable statistical effects. Future research could address these limitations by combining archival case comparisons with cross-country datasets on sovereign bond liquidity, private collateral creation, repo funding, credit booms, crisis outcomes, and institutional credibility. Comparative studies of emerging markets would be especially valuable because fiscal institutions, monetary credibility, and legal enforcement may be more uneven in these economies.

The overarching lesson is that the trustworthiness of safe assets depends heavily on the reliability of the institutions backing them. Financial stability therefore requires more than the availability of liquid claims. The governance architecture is indispensable as it preserves the conditions under which those claims could continue to circulate without becoming objects of suspicion.