MIT OpenCourseWareLecture 3: Distributed Ledger as a Solution to an Information Problem
MIT OpenCourseWare
Course Outline and Motivation 0:00
Robert Townsend opens the third lecture by framing distributed ledgers as a solution to an information problem tied to fragmented markets. He explains that central banks often pursue distributed ledger use cases driven by what the technology can do rather than by a clear economic need, and he wants to correct that by starting from efficiency as the actual objective of public policy and private innovation. The plan is to take a standard economic setup, fragment it into pieces to create frictions, watch regulators try to fix the resulting problems, and then show that even fixing prices leaves an information problem that distributed ledgers can address.
Pareto Efficiency and Competitive Equilibrium 3:03
Using a simple exchange economy with two goods and two households, Townsend reviews Pareto efficient allocations, where you cannot make one household better off without making the other worse off, shown as a curve of tangent indifference curves with equal marginal rates of substitution. He shows how solving a planner's problem that maximizes a weighted sum of household utilities traces out this same efficient frontier as the weights change, and how a private ownership economy with endowments and trading prices produces a competitive equilibrium that lands on one particular efficient point. He then generalizes this to many goods, many households, and firms with production sets, defining the utility possibility frontier and the formal condition for Pareto optimality: no feasible allocation exists that helps someone without hurting anyone else. Competitive equilibrium is defined as a price and allocation where every firm maximizes profit, every household maximizes utility within its budget, and total consumption never exceeds what is available, with the key theorems stating that such equilibria are Pareto optimal and that any Pareto optimal allocation can be reached this way given lump sum wealth redistribution.
Common Prices and the National Market System 15:04
Townsend turns to the first problem from fragmented markets, the requirement that trading happen at a common price. He points to the United States, where multiple stock exchanges quote different prices, and describes the Securities and Exchange Commission's National Marketing System from 2005, which forces exchanges to route retail orders to whichever venue offers the best price. He notes a recent paper by Chester Spatt and a colleague, called Regulating Market Microstructure, on later amendments covering smaller pricing increments, access fees, and order transparency. He contrasts this with an antitrust case against Amazon, where Amazon defended its price matching and deal features as pro consumer, and observes that this stance actually runs against the spirit of the National Marketing System's uniform pricing standard, leaving him to wonder aloud why financial market regulation seems to demand something that regulation of other markets does not.
The Ostroy Starr Information Problem 21:30
Townsend introduces the 1974 Ostroy-Starr paper on the information problem of decentralized exchange, where traders are exogenously paired off and forced to trade at common prices, removing price disagreement as an issue while still exposing frictions from fragmentation. In this setup traders hold endowments of goods or assets, are matched in a known sequence of pairwise meetings, and must trade so that the value given up equals the value received in each swap, with the shared goal of reaching the efficient Walrasian allocation despite never all meeting at once. The real question is what information each trader needs at each meeting to make this possible, and Townsend lays out increasing levels of decentralized information, from knowing only current excess demands and running totals, to knowing trading partners' full histories, up to a fully centralized case where every trader knows everyone else's trading history. The formal theorem shows full centralized information guarantees reaching equilibrium in a single round of trading, while the richest decentralized information alone cannot guarantee this for every possible economic environment, a result Townsend calls the twin impossibility theorem.
Money as an Institutional Workaround 35:00
Responding to a student's question, Townsend clarifies that in this model traders are not maximizing their own utility during trading, they are simply following a designed rule aimed at achieving the socially agreed efficient outcome, with individual preferences mattering only in defining that target. He then introduces the first institutional workaround to the information problem, the use of money as a medium of exchange, meaning a single chosen good that every trader holds in enough quantity to buy what they need regardless of meeting order. Formal theorem four shows that with this money good in place, even the weakest information level, which ignores both trader identity and trading history, suffices to complete all trades in a single round, provided each trader holds enough of the money good to cover their positive excess demands. Townsend notes this still requires picking the right good as money, and closes by flagging that without the quid pro quo condition the problem would be far easier, setting up further workarounds involving a broker dealer and an overdraft facility for the next part.
Liquidity Saving Mechanisms in Payment Systems 41:31
Large value payment systems run by central banks settle claims between commercial banks and move sums equal to about one hundred times US GDP. These systems once settled at the end of the day, which proved risky when banks failed to produce funds, so they shifted toward real time gross settlement. Because instant settlement would force banks to hold enormous liquidity in escrow, banks and central banks developed liquidity saving mechanisms, such as the Bank of England system, which queue payments and let banks receive incoming funds before releasing their own, reducing the total liquidity needed. Banks still act strategically, waiting until late in the day to submit payment orders to conserve liquidity, a workaround that would be unnecessary if all trades were already on a shared registry.
A Single Broker Dealer as Workaround 48:01
Another institutional fix imagines designating one large trader as the sole intermediary, a kind of giant warehouse, through whom all trades pass. This dealer must hold large enough inventories of every good or security to satisfy whatever demands other traders bring, regardless of who arrives first. The catch is that this concentrates market power in one actor who could set prices like a monopolist, undermining the very theorem that promised competitive outcomes without money.
Concentration in Payments and Foreign Exchange 51:01
Real world payment processing shows heavy concentration: three firms, Fiserv, Jack Henry, and FIS, dominate US bank clearing, with Fiserv alone handling 42 percent of the market. A similar pattern appears in foreign exchange, where broker dealers like Citibank, Deutsche Bank, and UBS handle trading in dominant currencies such as the dollar, yen, and pound, while other currency pairs, say Brazilian and Colombian money, must be routed through dollars. Core dealers absorb exchange rate risk while peripheral dealers treat currency as a hot potato, passing it along quickly, and exchange rates end up tied to the capital positions of just a few core banks. A distributed ledger might allow more currencies to trade directly and reduce this concentration.
Overdraft Credit and Medieval Trade Fairs 56:30
A third workaround gives traders overdraft credit facilities they can spend and must repay by value at the end of trading, effectively creating money that can be issued in whatever amount is needed since it will be settled later. Medieval French trade fairs worked this way: bankers kept ledgers, traders deposited coin for an IOU, and goods were traded in a set sequence, cloth, then leather, then spices, with settlement at the end. Coins varied and were periodically debased by kings melting them down, yet the unit of account on the ledgers, called ghost currency, persisted even after the physical coin changed, showing that the unit of account is a separate function from the store of value or medium of exchange. When traders could not repay their overdraft, they issued promissory IOUs called lettres de foire, letters of the fair, which later became circulating notes and eventually contributed to financial crises.
Default Risk and the Repo Market 1:03:00
Overdraft systems carry default risk, and this showed up starkly in the US repo market around 2008 and 2009, when broker dealers were found to be extending intraday credit larger than the entire US monetary stock, prompting reforms to change settlement timing so dealers carried more balance sheet coverage.
The Ostroy-Starr Counterexample 1:04:33
The impossibility theorem shows decentralized information alone cannot achieve efficient trade without a centralized ledger. Ostroy and Starr proved this by constructing a counterexample: two different economies where, after several rounds of pairwise trading among five traders, the traders cannot tell which economy they are actually in, leaving them without enough information to know what to trade. A former student's attempt to verify this example ran into trouble, and Ostroy and Starr themselves could not confirm it either, though the underlying logic still illustrates why partitioned trade without a shared ledger fails.
Distributed Ledgers Solving Real Tracking Problems 1:12:34
Recording trades on a shared ledger that everyone can see solves the information problem the counterexample points to, and real applications already show this. Walmart Canada uses a distributed ledger tracker to resolve invoice and payment disputes in shipping. De Beers tracks diamond origin on a distributed ledger. Ghana has built a trade finance prototype, working with Singapore, so small exporters can be assured of payment in a foreign currency once goods are verified as shipped, with funds held in escrow through a smart contract, a topic saved for the next lecture.
AI-generated summary. It can be wrong or incomplete - check anything that matters against the original.
