The U.S. Securities Market at Independence, the Civil War, and the Rise of Big Business: The Development of Microstructure

Richard Sylla
New York University


An earlier paper by G. D. Smith and me showed that the United States almost from its inception was at the forefront of nations in relative equity holdings and stock market development. Goldsmith's comparative national balance sheets indicate that during the first seven to eight decades of the 19th century the United States had a higher ratio of corporate stock both to national assets and to financial assets than did Great Britain or any of the other countries for which data were available From the 1870s to World War I, when London unquestionably was the world's preeminent financial center, Britain caught up with and passed the United States in terms of both ratios. Subsequently, the United States regained its lead over Britain on these scores during the 1920s boom, lost it during the Great Depression, and again regained it after mid-century.

Securities markets came early to both Britain and the United States. Securities were traded in London at a variety of places - the Bank of England, the Royal Exchange, and City coffee houses - from the late 17th century. In 1773, one of the coffee houses, New Jonathan's, posted a sign over the door reading "The Stock Exchange," and brokers drew up rules of admission, conduct, and cost sharing. By 1801, trading activity outgrew the coffee house; members raised a capital and built a dedicated Stock Exchange building, which opened in 1802.

Developments in the United States were similar, but were compressed into a much briefer period. Informal securities trading began in the era of the War of Independence, which led to an outpouring of paper claims. Corporate banks and other companies with tradable debts and shares appeared in the 1780s. Hamilton's funding of U.S. debts into uniform long-term bonds and his Bank of the United States together brought some $80 to $90 million of pristine new securities to the markets during 1790-1792. Philadelphia brokers organized a "board" in 1790, followed by New Yorkers in 1792 in their celebrated Buttonwood Agreement. The New Yorkers dedicated the Tontine Coffee House in Wall Street as their exchange in 1793. In 1817, they formed the New York Stock & Exchange Board with its own constitution and bylaws; its name was changed to the New York Stock Exchange in 1863.

This paper considers the microstructure of the U.S. securities market during the "long" 19th century, that is, from its inception around 1790 to World War I, when the United States became a world leader in finance. The questions examined are: What is meant by the "microstructure" of a securities market? What was the U.S. microstructure at particular points of time? How and why did it change over time? In particular, how did it change historically in response to securities market growth? To changes in the degree of competition? What was the impact of technological change on microstructure? And how was microstructure affected by laws and regulations? Economic and financial historians are just beginning to explore these questions.

Elements of Microstructure

Modern discussions of microstructure point to the following related elements:

Transactions costs. How much does it cost a market participant to buy or sell a block of shares or bonds? In most organized financial markets, e.g., securities exchanges, these costs are and have been low in relation to the values of the assets traded compared to transactions costs in non-financial markets. But they can be high relative to annual dividends, interest, or returns, thereby putting a damper on active trading. Moreover, if transactions costs charged are greater than actual economic costs, there is a deadweight loss as some potential market participants choose not to trade. In addition, there are costs of simply having, say, a stock exchange to facilitate transactions. How these costs are covered can have a powerful impact on microstructure.

Price discovery. By what processes are securities prices determined? How close are they to equilibrium values? Modern securities markets are thought to be as close as humanly possible to the ideal perfect market. But they are not perfect even now, in part because information is less complete or perfect than it could be. Why? Because market participants who completely revealed their trading intentions could be put at a disadvantage.

Informational transparency. Securities markets produce not only transactional opportunities, but also information, e.g., a price quotation, that has value to actual and potential traders. How accurate is this information? Who controls it? How quickly is it conveyed to potential users? At what cost?

Order flow consolidation. A market transaction occurs at both a particular place and at a particular time. To the extent that more transactions can be done at the same place and the same time, the greater should be both the ease of trading for individual market participants and the accuracy of price discovery. Historically, these considerations go far toward explaining why securities markets produced stock exchanges, and why call markets often preceded continuous trading markets.

Liquidity. Buyers and sellers may have their own liquid funds. But if those were the only sources of liquidity, the ability to buy and sell whenever potentially profitable opportunities arose would be limited. Trades would be fewer, price volatility greater, and user costs of capital would therefore rise. These negatives can be avoided or reduced if buyers and sellers have access to credit or other outside sources of liquidity to finance their market operations.

Volatility. Modern studies of both short- and long-term price movements in securities markets conclude that price volatility is excessive, i.e., greater in the short-run than would be expected under a random walk, and greater in the long-run than would be expected given the so-called fundamentals of security price determination, e.g., earnings, dividends, and their growth. Again, excess volatility raises the costs of capital to its users above what it would be if volatility were in some sense normal.

Immediacy. Call markets - trading in a security at set times when it is "called" and offers to buy and sell are then made, gave way historically to continuous markets so that a security could be bought or sold at any time the market was open. When trading volumes were much less than they have been in recent decades, call markets probably made sense as a means of realizing the advantages of consolidating the order flow, at the sacrifice of immediacy. Higher trading volumes brought immediacy, but possibly at some cost in terms of short-term price volatility. Interestingly, one aspect of today's large trading volumes, namely institutional large-block trading, has caused renewed interest in call markets that can be carried on electronically and could possibly reduce price volatility by consolidating order flow.

Functional specialization. To a great extent, the development of securities markets has brought about a high degree of functional specialization. Markets that began as meetings of buyers and sellers evolved into markets of buyers, sellers, brokers, dealers/jobbers, specialists, floor traders, "two-dollar" brokers, odd-lot brokers, and so on. An issue of regulation is whether some forms of specialization should be required if they do not evolve "naturally." The London Stock Exchange required one to be either a broker or a jobber/dealer/specialist, but not both, because of a perceived conflict of interest. A broker's duty is to get the buyer or seller the best price. A jobber's self-interest is to get himself or herself the best price, which could be the worst price for a seller or buyer. In U.S. markets this way of thinking never caught on, although it was present in the earliest days (see below). In America, one could be both a broker and a dealer. Under what conditions would this distinction, or lack thereof, produce distinct market outcomes?

Historical Microstructure: Three Key Eras

Although some elements of the microstructure of the first U.S. security markets, e.g., liquidity, developed more or less continuously over time, major changes occurred in three historical periods when new and greatly enlarged demands fell upon them. These periods were those of independence (here meaning particularly the era when the Constitution was adopted and implemented in part by means of Hamilton's financial program), the Civil War decade, and the rise of big business in manufacturing, roughly from 1885 to 1915.

Independence. Hamilton's policies brought some $70 millions of Treasury bonds and $10 million of Bank of United States stock to the markets, such as they were, in the first years of the 1790s. Other stocks and bonds of banking, insurance, transportation, and utility companies also appeared at this time. From the start, New York was the center of speculation and active trading. Initially, securities were sold at public indoor and outdoor auctions. A set of trading rules put forth in September 1791 by leading dealers, brokers, and auctioneers has survived. The rules prohibit auctioneers from trading for their own account and, interestingly, they appear to embody the English functional separation of brokers from dealers/jobbers at the auctions. This seems to be the first and last time that the English distinction was applied in U.S. markets, most likely because the volume of trading during the early period soon became too small to sustain it. By 1800, a New York market leader, Nathan Prime, acted as both dealer/jobber and broker in the Wall Street securities markets. Combining the two functions, contrary to English exchange rules, obviously added liquidity and a version of immediacy to the thin U.S. markets of the period around 1800.

Intense speculative activity during 1791 and early 1792 led to a crash, Wall Street's first, in February and March, 1792. The state cracked down in April with a law banning public auctions of securities. This was a key historical regulatory development, for it forced New York's leading brokers in the Buttonwood Agreement of May 1792 to form a private club for securities trading. There were advantages to a club beyond coping with government regulation. The Buttonwood Agreement also pledged its signers to fixed minimum commissions of one quarter percent and to give preference to other club members in their dealings. This was the precedent that began the long history of the New York Stock Exchange in formulating club rules and regulations designed to promote the interests of its members even if they conflicted with the interests of other market participants and the utility of the securities market.

One important feature of early trading was the so-called time contract (or time bargain or wager) in which traders would agree on a price at which to exchange a specified number of shares or bonds at a specified future date. Often under such contracts no exchange of securities was actually made. When the specified date arrived, the contracting parties would simply compare the contract price with the current market price, and settle the difference in cash, much as happens in modern futures and other derivatives markets. Given the nature of these contracts, it is difficult to know their relative importance in all trading, but some scattered evidence indicates that they accounted for at least 20% of trading volume. Time contracts of this nature conserved liquidity at a time when the United States had few banks and was "young" in wealth accumulation. They disappeared during the 1840s and 1850s when margin trading based on enhanced market liquidity replaced them. Time contracts were not legally enforceable. This drawback enhanced the exclusive-club nature of the stock exchange; one needed to know a lot about the party with whom one traded and to have some club rules and penalties for non-performance as substitutes for the missing legal sanctions.

The New York Stock & Exchange Board constitution of 1817, like the earlier Buttonwood Agreement, came at a time of increased market volume. The federal debt was substantially enlarged during the War of 1812, and a new Bank of the United States with $35 million of capital stock, rather than the $10 million of its predecessor, was organized in 1816. The 1817 constitution reaffirmed minimum one-quarter percent commissions on securities transactions, formalized the call-market rules, registered all transactions made at the Board with this information available only to members, and authorized the Board secretary to provide weekly quotes to only one "price current" newspaper. The Exchange had discovered that the information it created had value, and its procedures for handling it favored members over considerations of informational transparency.

One of the two most important developments affecting microstructure between 1820 and 1860 was in the area of liquidity. The presence of the stock exchange allowed New York banks to provide lines of credit and liquid call loans to brokers, who in turn used these loans to finance their own and their customers' market operations. With active markets, securities could be collateralized up to 80 to 90% of their market value. Such lending opportunities with high liquidity meant that the New York banks could pay interest on, and thus attract, interbank balances from all over the United States. Banks outside New York often had surplus funds that they could hold as interest-earning reserves in New York, and they also found it convenient and profitable to maintain New York "exchange" for their customers when the country had no national paper currency. The symbiotic relationships that evolved in the antebellum decades between the U.S. banking system and the money-center New York banks, and between the New York banks and the city's securities markets were enshrined in federal law during the Civil War, when the National Banking acts recognized New York as the central reserve city of the U.S. banking system.

The other important antebellum development affecting microstructure was the telegraph. It greatly enhanced possibilities for arbitrage between the securities markets of different cities. The telegraph also made it more convenient and immediate to buy and sell securities at a distance. As with the market liquidity supplied by New York's banks, the New York securities market was a prime beneficiary of the telegraph. Orders that might have gone to nearby local markets were now sent to New York with no loss of immediacy, in order to take advantage of New York's broader and more active markets as well as its facilities for securities market credit. If there ever was any doubt about which city would become the financial center of the United States, it was erased well before 1860.

The Civil War Era. Between 1860 and 1866, the federal government's debt increased from $65 million to $2,800 million, much of it in negotiable securities. This influx of debt amounted to an approximate doubling of the amount of all American securities outstanding in 1860. The greenbacks, a wartime fiat currency, broke the link of the dollar to gold, thereby creating a speculative market in the yellow metal. New petroleum and mining securities also appeared on the markets during the war; small, specialized exchanges opened in New York to trade them.

The New York Stock Exchange, as it renamed itself in 1863, remained an elite club of members and securities; it kept its time-honored minimum commissions and kept out the new and unseasoned issues. Enlarged wartime trading volumes nonetheless created competitive threats to the NYSE. Two "Open Boards of Brokers," more open than the NYSE club, arose to challenge it for business in 1862 and 1863. The competition cut transactions costs; minimum commissions were an eighth percent or less on the Open Boards. Reluctantly, the NYSE cut its minimum to an eighth, and less reluctantly it put in a rule to expel any member who dealt on the rival boards.

In a clever move, the NYSE made its memberships saleable in 1868. This provided an inducement for the main Open Board to be merged into the NYSE in 1869, thereby eliminating the main competitive threat. The NYSE emerged as an exchange with 1,060 seats, roughly double the pre-merger level. In 1879, 40 more seats were added, and the number remained at 1,100 into the 20th century.

Limiting the number of saleable memberships was consistent with maximizing the value of a membership, but it would lead to competitive alternatives arising whenever market volume increased substantially. Entry to exchange membership was far easier and less costly in London, where the Stock Exchange had several times the membership of the NYSE. The NYSE's restrictive membership policies led to restrictive listing policies; it traded governments, financials, and rails, and not much else, because the other securities available for trading were new, unseasoned, less active, more volatile, and less convenient and rewarding for the 1,100 hundred members to handle. Hence, New York and the United States, unlike London and Great Britain, would not have one central securities exchange but rather a number of exchanges and non-exchange markets.

Two important communications developments came in the Civil War era. The first successful transatlantic telegraphic cable came in 1866, enhancing the possibilities for arbitrage between the United States and Europe, where many U.S. securities were owned and traded. And stock tickers were introduced in 1867, quickly spreading throughout the country. The advent of stock price tickers along with the larger post-merger NYSE and growth of trading volume soon gave rise to a major change in microstructure: continuous trading began in 1871, and gradually replaced call trading, which was finally abandoned for stocks in 1885 and for bonds in 1902. Immediacy was gained, but at some cost in consolidating the order flow.

Tickers, of course, also promoted domestic arbitrage between the markets of various U.S. cities. The NYSE, as the largest and most active market, tended to set the prices of the securities it listed for the whole country and even the world. International and domestic arbitrage, however, raised the spectre of non-members' free riding on the NYSE price discovery mechanism. The NYSE was caught in a dilemma. While enjoying enhanced prestige and greater business as a result of the communications breakthroughs, the NYSE also saw that non-members were benefiting from the information it created, and so it tried in various ways to impede the flow of information to them (see below).

The Rise of Big Business. Because of its exclusive-club nature, its sacrosanct fixed commissions, and its members' interest in maximizing the value of their seats, the NYSE was an institution almost designed not to be responsive to change. Focusing on the largest and most actively traded issues - financials, rails, and governments before 1900 - it was less than receptive to the securities of newer manufacturing, commercial, and extractive companies when these appeared in increasing numbers during the late 19th century. Moreover, it quoted prices as a percentage of par value and dealt in lots of 100 shares with a par value of 100 per share, and so the par value of the basic transaction was therefore a hefty $10,000. And the NYSE based its minimum commissions on par, not market, values. If a growing manufacturing company sought to attract buyers for its securities by issuing them at a par value below 100, or if its securities sold at a discount to par, these NYSE features discriminated against it, provided, of course, it could get a listing. Before the 1890s, it typically could not get a listing.

The "high end" focus of the NYSE, the high-volume trading of established companies' securities in relatively large transactions, created an opening for competitors that were more flexible in their listing and trading regulations. In response to new demands for corporate finance and the NYSE's lack of response, the Consolidated Stock & Petroleum Exchange was formed in New York in 1885. It cut commissions from an eighth to a sixteenth, and it dealt in basic lots of 10 rather than 100 shares. Besides providing an exchange for securities the NYSE would not admit, it provided an alternative market for NYSE-listed issues. The NYSE fought back, reactively. It established an "unlisted" department in 1885, for securities that did not meet its customary high standards, and kept it for a quarter century. It also tried to ban dual memberships in other New York exchanges, and it prohibited ticker and phone links (the telephone arrived in the late 1870s) between its members and non-member brokers, a prohibition later declared to be illegal. It even went so far as to interfere with the timely dissemination of its quotes, thus reducing opportunities for domestic and international arbitrage. It was in the nature of arbitrage that someone else, somewhere else might earn a commission or part of a commission based on NYSE quotations, but the NYSE found this sharing of what it considered its entitlement to be galling. So it acted to make its own and other markets less liquid and more volatile.

These countermoves in the face of competition, however short-sighted they might seem in some macro-efficiency sense, did help the NYSE to realize its own objectives. The Consolidated Exchange began to decline in the 1890s and went into eclipse by World War I. The NYSE listed a few industrial securities in the 1890s, and the list grew as this class gained respectability after the turn of the century. Thus, the NYSE was able to maintain its hold on the now larger and more diverse high end of the market, with less seasoned issues traded in the nearby, non-competing "Curb" market, and less actively traded issues handled by the over-the-counter dealer market.

The final move to modernity of the pre-World War I era came in 1915, when the NYSE changed from quoting and trading on a percentage-of-par-value basis to one based on actual dollar prices. By that time, it is evident, most aspects of the microstructure of U.S. securities markets had taken the forms they would hold for another half century, when once again new demands were placed on them, competitive challenges again arose and were resisted, and out of it all new microstructures resulted.