TRANSACTION CHARACTERISTICS

In a marine insurance contract, an insurer (“underwriter”) assumes some of the maritime risks on a vessel or cargo, or both, in exchange for a premium. The risks covered may include a variety of risks at sea or in port, for a particular voyage or for a period of time. Marine insurance originated in the Italian city states in the middle ages. By the start of the eighteenth century it had become a familiar practice and, because of the hazardous nature of maritime commerce, a crucial input for trade. However, insurers had to contend with numerous sources of uncertainty and complex information asymmetries that created agency problems for both the insurer and the insured. These can be separated into three main categories.

First, the probability of a ship or its cargo being lost or damaged (and therefore, the appropriate premium) depended on numerous risk factors including the route, the season, the age and seaworthiness of the ship, and the quality of its crew and armament, as well as, in wartime, the danger of capture by enemy naval vessels and privateers, and the possibility of seizure in a foreign port. Some of these factors could be taken into account by the underwriter when determining the premium. Often, however, merchants had better information than underwriters about various aspects of the risk. Thus, one underwriter complained that “with the keenest penetration and judgement, it will rarely happen that [the underwriter] is on an equal footing, as he ought to be, with the insured.”¹

Second, there were many opportunities for moral hazard on the part of the insured. These ranged from excessive risk-taking, such as sending unseaworthy ships to sea or attempting to carry goods into a blockaded port; to outright frauds, such as deliberately sinking an insured ship, misrepresenting the value of the goods, insuring the same goods multiple times, or seeking to insure a ship already known to have been lost.

For example, it was perfectly legal, and not uncommon, for merchants to insure the same goods with multiple policies. Legally, the merchant could not recover in total more than the value lost.²

Legislation passed in 1746 sought to control moral hazard and gambling by making it illegal to insure without an insurable interest, and banned reinsurance (except in special circumstances) because there had been instances in which underwriters had fraudulently sought to reinsure their interest after receiving private information of a loss.³ However, the possibility of fraudulent overinsurance and deliberate shipwreck remained a serious concern for underwriters. In 1816, for example, a merchant took out eight policies with brokers in London, Hull, Glasgow, and Dundee to insure a ship and goods worth about £1,850 for a total of £5,405, and then deliberately scuttled the ship after secretly removing some of the cargo. By chance, the fraud was discovered, and the culprit narrowly escaped a death sentence.⁴

Third, the financial stability of the underwriters was a major concern for those buying insurance. Private underwriters might become bankrupt, die, emigrate, or abscond before a claim could be made, and their financial stability was frequently uncertain, especially in wartime, when the increased risks both to ships at sea and to the fortunes of the underwriters made private underwriting more precarious. In addition, because both the timing and route of voyages in wind-driven vessels were inevitably unpredictable, and because assembling witnesses and documentary proof to support a claim in court could be a slow and costly process, underwriters had frequent opportunities to contest claims or delay payment. As a result, merchants preferred to insure with underwriters they perceived as nonlitigious and financially secure.

In the context of these information asymmetries and agency problems, information was critical. Underwriters needed access to prompt and accurate intelligence on the movements and condition of particular ships, the “character” of the merchant being insured and the captain of the vessel, and on political developments at home and abroad, as well as the experience to weigh this information correctly to determine premiums. In the words of one contemporary underwriter,

Information, however, traveled slowly in the eighteenth century. The latest news from foreign ports was frequently months old. Ships sometimes disappeared without a trace, and if a ship was lost or damaged, it was frequently costly, and sometimes impossible, to verify or disprove events that had occurred in distant ports or at sea. It could also be difficult for merchants to accurately judge a prospective underwriters' prudence and financial solidity, or for an underwriter to learn about a merchant's “character”; personal connections and the “trust” embodied therein were a valuable asset for any businessman.

Despite all these difficulties, marine insurance provided crucial support for the expansion of trade throughout the eighteenth and nineteenth centuries. The market was able to function because institutions were created to mitigate the information asymmetries and agency problems inherent to marine insurance at the time. What were the important characteristics of these institutions, and how did they arise?

MARINE INSURANCE IN BRITAIN, 1720–1844

At the start of the eighteenth century, marine insurance in Britain was carried out entirely by private individuals.⁶

Many underwriters were merchants who wrote insurance on the side, but any wealthy individual willing to dabble could underwrite a policy. A merchant wishing to insure generally employed a broker to draw up a policy and present it to private underwriters for their signature. If they considered the premium acceptable, the underwriters wrote their name on the policy, along with the amount they were willing to insure.⁷ Risks were usually shared among several underwriters. Although there was as yet no single center for underwriting, merchants and underwriters generally frequented several coffee houses around the Royal Exchange and near the Thames, where they exchanged news and gossip, and transacted marine business including ship auctions and marine insurance.⁸

During the stock market boom that culminated in the “South Sea Bubble” of 1720, several groups of merchants and speculators began petitioning to obtain charters for joint-stock marine insurance corporations.⁹

The promoters of the proposed corporations argued that they would provide cheaper and more secure insurance than the existing system of private underwriters. Private underwriters had unlimited liability for losses, but there had been instances of failures, especially in wartime. In contrast, the proposed corporations would be backed by a large capital fund, and in the event of a claim, the merchant could more easily recover his losses from a corporation than from many individual underwriters separately. In addition, corporations would expand the pool of capital available for underwriting by enabling those without specialist knowledge of marine risks, or with relatively modest amounts of capital, to act as insurers by entrusting their underwriting decisions to experts.¹⁰

The proposed charters were opposed by merchants and private underwriters in London and Bristol, who claimed that the existing system was adequate, and that a monopoly would harm trade. It was widely assumed on both sides that if charters were granted, the proposed corporations would rapidly drive the private underwriters out of the market.¹¹

The argument was settled when the two main groups of promoters offered the King £600,000 (to pay off the debt on the Civil List) in exchange for charters. Two joint stock corporations (the Royal Exchange Assurance and the London Assurance) were subsequently incorporated as part of what later became known as the “Bubble Act” of 1720. The Bubble Act made it illegal for joint-stock companies to operate without a corporate charter. In all industries except marine insurance, however, other kinds of unincorporated companies, including partnerships and trusts, were still allowed, and businessmen were later able to use these devices to create (highly imperfect) substitutes for the joint-stock business corporation.¹²

However, marine insurance received special treatment: all firms and partnerships, apart from the two corporations chartered by the Bubble Act, were barred from writing marine insurance (crucially, however, private underwriting by individuals was still allowed).

The Bubble Act had a tremendous impact on the development of marine insurance in Britain until the relevant clauses were repealed in 1824, but its ultimate effect was neither foreseen nor intended in 1720.¹³

Contrary to expectations, private underwriting not only survived, but flourished. In 1810, it was estimated that “both [corporations] added together would not exceed what two of the most considerable individual underwriters would write in one year.”¹⁴

Why, despite their advantages, did the two chartered corporations fail to dominate the British marine insurance market? In the remainder of this section, I describe how the market developed. Later, I argue that the way in which the market developed—in particular, the way in which the private underwriters managed to overcome the information asymmetries and agency problems described in the previous section—meant that the two chartered corporations faced a lemons problem that hindered their expansion.

Edward Lloyd's coffee house opened for business in the mid-1680s, and by the turn of the century, had become a center for ships news and other activity connected with shipping.¹⁵

By the 1720s, it had emerged as a focal meeting place for marine insurance. Merchants and brokers could get policies underwritten there quickly and conveniently because of the number of underwriters in attendance, and underwriters in turn were lured by the prospect of plentiful opportunities to underwrite. In this way, long before it gave rise to any formal association, Lloyd's coffee house became the center of the marine insurance market in London.¹⁶

In order to compete with the corporations, the private underwriters had to overcome the three kinds of information asymmetries and agency problems discussed previously: collecting and interpreting the information needed to determine the correct premium; overcoming moral hazard problems such as fraud on the part of the insured; and generating confidence in the good faith and financial security of the underwriter. Business practices at Lloyd's, which emerged and evolved gradually over an extended period of time, managed to partially overcome each of these difficulties.

First, to meet his customer's needs for shipping news, Lloyd made a systematic effort to gather and disseminate the most accurate and up-to-date information on ship movements, political developments, and any other relevant information. Some of the news came from the customers themselves; but Lloyd also employed runners who went along the docks picking up news of arrivals, departures, losses, and other relevant gossip, and relaying this information to the coffee house. He began to build up a network of correspondents who were paid to send him shipping information from domestic and foreign ports, and cultivated a special relationship with the post office (which had its headquarters near the coffee house): in exchange for an annual fee, letters addressed to Lloyd's were carried free of postage, sorted specially and held for collection by a Lloyd's messenger. The combined effect of these efforts was to give Lloyd's a “practical monopoly of complete and up-to-date shipping intelligence.”¹⁷

As the news was brought in, it was announced publicly from a pulpit built for the purpose. From 1692, Lloyd also published a weekly newspaper of shipping news (which later became Lloyd's List).¹⁸

In the early 1760s, a committee of Lloyd's underwriters began employing surveyors to assess the condition of ships, and compiling this information in a register of ships that became known as Lloyd's Register.

Because interpreting all this information to determine premiums involved weighing numerous factors, underwriting was ultimately a matter of individual judgment, requiring experience and familiarity with the routes, vessels, people, and circumstances involved. Lloyd's attracted a wide variety of underwriters, many of whom were active or retired merchants, and whose specialist knowledge of particular kinds of risks enabled them to make these kind of judgments. This was particularly important in evaluating cross risks (voyages between two foreign ports). As one broker explained:

Once a policy had been begun by a respectable “lead” underwriter with the appropriate expertise, other underwriters, despite lacking specific knowledge of the risk, were often willing to “follow the lead” and underwrite at the same premium.

Second, there was the problem of moral hazard on the part of the insured. The concentration of business at Lloyd's may have helped to control some kinds of fraud, including fraudulent overinsurance, by raising the probability of detection. In addition, contemporary accounts suggest that the trust generated through repeated interaction between merchants and brokers, and between brokers and underwriters, was key in helping to control (though not eliminate) many kinds of fraud. Here, the broker's role was crucial.²⁰

A reputable broker, such as J. J. Angerstein, was “very careful with whom I open accounts,” and stated that “other brokers as well as myself refuse business continually” from those he did not consider “respectable.”²¹

Underwriters would therefore underwrite policies more readily, and at a lower premium, for a reputable broker than for someone they did not know or trust, and endeavored to maintain a reputation for honest and open dealing, so as not to risk losing the business the broker had to offer by underwriting imprudently or appearing overly “litigious.” They were even often willing to pay claims in cases where an unanticipated change in the voyage or a mistake on the policy might have rendered the policy technically invalid.²²

Again, however, the broker's role was key. Angerstein noted that “the private underwriters will settle the loss for a man of character, where they will not for a man whom they suspect.”²³ Another broker, who had occasion to settle heavy losses in 1809, stated that “it was settled on the letters only of the different merchants, and on the confidence the underwriters placed in me.”²⁴

Because reputable brokers could get policies underwritten more quickly, at lower premiums, and with better underwriters than the merchants could achieve independently, merchants also valued their relationships with good brokers, and had an incentive to avoid suspicious dealings lest they jeopardize these relationships. Thus, all market participants had an incentive to refrain from opportunistic behavior and maintain a reputation for prudence, fair dealing, and respectability.

There remained the problem of the underwriter's security. Although private underwriters did sometimes fail, such failures were surprisingly rare, even during the Napoleonic wars, when several calamitous events produced large claims. The reputation mechanism played a role here:

The terms of credit granted to brokers in the collection of premiums also helped to ensure the underwriters' security. Underwriters granted brokers generous terms of credit, often over a year, to pay premiums.²⁶

This meant that at any time the private underwriters had, in effect, two sources of capital: their own private fortune, with full unlimited liability; and a fund of premiums owed to them by the brokers for policies they had already underwritten. In turn, the brokers often allowed credit to the merchants, so that in effect, the premium was often not actually paid until after the risk had been run. Even if an underwriter became insolvent, there were generally sufficient outstanding premiums due from the brokers to pay off any claims that might arise on outstanding policies.²⁷

Thus, by the late eighteenth century, business practices that emerged at Lloyd's had managed to partially overcome each of the three kinds of information asymmetries and agency problems described previously by collecting and making efficient use of shipping information, while (imperfectly) protecting underwriters against fraud, and merchants against insolvency and opportunism on the part of the underwriters. Remarkably, these practices emerged despite virtually no conscious collective efforts to design new rules or change old ones. Rather, Lloyd's was “a striking example of evolution as distinguished from creation. Conditions have been made, rules instituted not in preparation for new factors and developments, but to systematise a practice which had already been adopted to meet the requirements of commerce as they arose.”²⁸

Yet the system worked so well that Lloyd's became the most important center of marine insurance in Europe.

The evolving legal environment both reflected and reinforced the dominance of Lloyd's. Until the mid-eighteenth century, insurance cases had been decided for the most part in an unsystematic, ad hoc way. However, Lord Mansfield, who was Chief Justice of the Court of King's Bench from 1756–1788, took major steps towards rationalizing and setting out legal principles of insurance. The principles and precedents that Mansfield established were generally derived from mercantile practice and custom, including the principle of “utmost good faith,” according to which any misrepresentation or concealment of facts by the insured, or deviations from the planned voyage without reasonable cause, would void a policy. Mansfield also simplified legal procedure, in particular, by eliminating the necessity for an insured party to bring legal actions against each underwriter separately in the event of a disputed claim (the “Consolidation Rule”). This strengthened private underwriting by removing one of the corporations' major advantages.²⁹

The Revolutionary and Napoleonic wars (1793–1815) substantially disrupted trade and increased the risks of international commerce.³⁰

The risk of capture increased the demand for marine insurance and drove up premiums.³¹ On average, the high wartime premiums more than compensated underwriters for the increased losses, and although some underwriters failed, many others made a fortune. The marine insurance market expanded rapidly as a result. The number of subscribers to the society of Lloyd's grew from 179 in 1775 to more than 2,000 by 1801.³² The two chartered corporations also prospered as their premium income grew (see Table 1), but the market remained overwhelmingly concentrated at Lloyd's.

This growth in business greatly accelerated the process of formal institutional development at Lloyd's. An ad hoc committee had been meeting sporadically since 1769, mainly to deal with leasing and furnishing suitable premises, but beginning in 1794, the committee's role expanded to encompass communications with the Admiralty and other branches of government regarding naval developments, convoys, and stamp duty, as well as prosecuting several instances of fraud. In 1796, it began to hold regular meetings and to issue an annual report. Since the 1770s, the committee had been making sporadic efforts to coerce those frequenting the coffee house to pay a subscription fee, but in April 1800, in response to overcrowding, they finally took effective measures to enforce the exclusion of nonsubscribers from the subscribers rooms.³³

In wartime, circumstances often changed rapidly, so prompt and accurate shipping, political, and military news was more important than ever. All the correspondence had previously been handled by the coffee house masters, but as the volume of correspondence grew, this arrangement became unsatisfactory, so in 1804 a secretary was appointed to manage it. The secretary further developed and extended Lloyd's' intelligence-gathering apparatus, for example, by appointing new correspondents abroad and subscribing to foreign newspapers. In 1792, Lloyd's had 32 correspondents in 28 ports. By 1820, the system had been formalized, and 269 “agents” had been appointed.³⁴

In 1811, a dispute arose over losses in the Baltic, and it became clear that the lack of clear regulations governing the operations of the committee and the use of information were the root cause of the problem. In response, a Trust Deed was drawn up that formalized the committee's responsibilities, including the supervision of the intelligence system, the appointment of agents, and the election of subscribers, and empowered the committee to act on behalf of the underwriters as a group.³⁵

Thus, in response to temporary needs caused by the increased demand for marine insurance during the Revolutionary and Napoleonic wars, Lloyd's had developed a formal governance structure and membership, and had formalized and strengthened its mechanisms for information gathering and self-regulation (however, underwriting remained strictly on an individual basis, and there was as yet no formal attempt to ensure the financial security of underwriters).

In 1824 the relevant sections of the Bubble Act were repealed, paving the way for an influx of new joint-stock corporations. As in 1720, it was widely expected that this would mean the end of private underwriting. One contemporary underwriter, for example, believed that “if companies are sanctioned, individual underwriting must cease, as has been proved by the example of all other countries, more particularly by the recent example of America.”³⁶

By the 1820s, however, the boom years were over, premiums were low, and the number of subscribers at Lloyd's was in decline. The average annual premium income on marine policies at the London Assurance, for example, was over £216,000 between 1810 and 1814, but less than one-tenth as much in 1820–1824 (Figure 1).³⁷ The Royal Exchange fared even worse: premium income plunged from £545,290 in 1814 to £14,562 in 1824, and whereas it had made average annual profits of £32,500 in 1793–1813, it lost an average of over £20,000 in 1814–1834.³⁸ In this inhospitable environment, many of the new companies failed, though several survived and made substantial inroads into Lloyd's share of the market, capturing as much as half the market by 1844.³⁹

In 1844 the Joint Stock Companies Registration and Regulation Act enabled companies to obtain all the privileges of incorporation by registering and meeting various legislative requirements, leading to a second large influx of new corporations.⁴⁰

Again, many of the new companies failed. Lloyd's dominant position in the market continued to come under threat from the companies, particularly during the 1870s, forcing it to evolve in significant ways. However, in the late nineteenth century, the flexible, market-based institutional structure of Lloyd's proved more adept than the companies at identifying and exploiting new underwriting opportunities, and it moved significantly into fire and accident insurance while remaining a major force in the marine insurance market.⁴¹

MARINE INSURANCE IN AMERICA, 1720–1844

In the United States, as in Britain, the Revolutionary and Napoleonic wars had a decisive and lasting effect on the development of the marine insurance industry; but unlike Britain, private underwriting in the United States had virtually disappeared by 1815, and the business was carried on almost entirely by joint-stock corporations.⁴²

In the early eighteenth century, American merchants generally obtained insurance in London, although the distance created considerable uncertainty and inconvenience. For example, orders for insurance on voyages to Britain were often sent ahead on ships expected to arrive before the ship being insured. If the ship made a fast passage and arrived before the orders for insurance, then the premium was sometimes saved, but this could also be precarious, as illustrated by the case of an agent in London who received instructions to insure a consignment of tobacco only an hour before the news of the ship's loss arrived, leaving the tobacco uninsured.⁴³

As the trade of the colonies expanded, an increasing amount of insurance was obtained locally. By the 1740s insurance brokers were operating in Boston and Philadelphia.⁴⁴

Local underwriters had better information on local risks, and their proximity enabled quick payment in case of loss, facilitated resolution of disputes, and avoided the necessity for merchants to rely on agents to represent their interests.

For example, in the early 1740s, Boston merchant Benjamin Dolbeare regularly obtained insurance in London on voyages between Boston and the West Indies. However, when the sloop Tryall was captured by the Spanish in 1739 on a voyage from Curacao to Boston, it took over a year to obtain payment from London. In 1742, he wrote for insurance on the Friendship, which was to sail to Antigua and then, if a war with France had not broken out, to Guadeloupe, and otherwise, to the English or Dutch Leeward Islands, and if markets there were unfavorable, to Jamaica, before returning to Boston. In the event, however, the vessel went no farther than Antigua and Dolbeare spent the next two years trying to prove this in order to obtain a partial return of premium, apparently without success. In March 1745 he settled his account with his London agent and subsequently rarely obtained insurance in London except on transatlantic voyages. Most likely, he had begun to insure his West Indies ventures locally.⁴⁵

Other merchants remained suspicious of local insurance and continued to insure in London, where insurance was generally cheaper, and the underwriters were regarded as more financially secure. In 1743, writing to his London correspondent for insurance, Philadelphia merchant William Till observed:

Although colonial America had nothing that could compare with Lloyd's, several active centers for private underwriting emerged. In Philadelphia, underwriters frequented the London Coffee House in the 1760s, and made several, apparently abortive attempts at organization.⁴⁷

In 1787, the broker Benjamin Fuller was able to obtain 34 underwriters on a large policy, “All the underwriters in the City (except 2 or 3) having Subscrib'd.”⁴⁸ Similarly, there were two coffee houses that functioned as meeting places for marine business in New York.⁴⁹

The Revolutionary and Napoleonic wars created enormous risks and opportunities for American merchants.⁵⁰

Britain's naval superiority effectively closed off direct communications between France and its colonies, so at the outbreak of war, the French opened their colonies to trade with American vessels. Under the “Rule of 1756,” Britain refused to recognize the neutrality of vessels trading directly between France and its colonies, but American vessels were able to carry on the trade indirectly by importing goods from French, Spanish, and Dutch colonies to the United States, where they paid customs duties, thereby conferring neutral status on the goods so that they could be re-exported to France and the continent; while in the reverse direction, European goods flowed indirectly to the colonies.

As a result of these new opportunities, and of protective tariffs and regulations that had been introduced by Congress in 1789, American trade volumes and shipping tonnage grew rapidly (Figure 2). In the words of one contemporary American observer:

At the same time, however, this growing trade was subject to considerable risk. Although America was neutral, both belligerents employed their navies and commissioned privateers to seize food bound from America for the other, and to seize enemy property carried in American vessels. The decrees regarding neutrality issued by the belligerents changed with sometimes extraordinary frequency, and in any case were not always carefully observed.⁵²

Six hundred American vessels were seized or detained in British ports between 6 November 1793 and 28 March 1794 alone, and crucially, British insurers could not be held legally liable for losses on neutral vessels captured by British ships.⁵³ In 1796–1797, when the French began to capture American vessels in the Caribbean in retaliation for Jay's treaty between the United States and Britain, the premium on one-way voyages to the West Indies rose from a range of 3–6 percent in the fall of 1796 to as much as 15–20 percent in the summer of 1797 and reached 25 percent in 1798, before American naval victories in 1799 and 1800 brought rates back down.⁵⁴

The growth in trade during the 1790s, coupled with its uncertainty, created the conditions for a rapid expansion of the American marine insurance market but placed considerable strain on the existing loosely organized system of private underwriting through brokers. However, independence had freed Americans from the Bubble Act's restrictions on the formation of joint-stock corporations, and after a period of political debate and uncertainty, the newly independent American states had begun to charter corporations for various purposes with increasing frequency during the late 1780s and early 1790s.⁵⁵

Thus, conditions were ideal for market entry by corporations.

The first American marine insurance corporation, the Insurance Company of North America (ICNA), was formed in Philadelphia in 1792 and chartered in 1794. The President and Directors were all leading underwriters, and most of the board were Philadelphia merchants. In contrast to Britain, where the two chartered corporations held a monopoly, no monopoly was sought in America. Philadelphia's remaining private underwriters initially opposed the INCA charter, but having observed its early success, they founded the rival Insurance Company of the State of Pennsylvania, which was also chartered in 1794.

Boosted by high wartime premiums, the Philadelphia corporations quickly proved highly profitable, and this in turn encouraged the formation of incorporated companies elsewhere: in Maryland in 1795, in Connecticut in 1797, in New York and Virginia in 1798, in Massachusetts and Rhode Island in 1799, and in Maine in 1800.⁵⁶

By 1810 private underwriting had virtually disappeared in the United States, and it was reported that “in every part of America the Insurances are done by incorporate companies,” including seven in Boston, eight in Philadelphia, five in Baltimore, and six in New York.⁵⁷ At the same time, the volume of American business insured in London fell sharply; in 1810, one American merchant in London estimated that “nineteen twentieths” of his consignments from America were insured in America.⁵⁸

The subsequent development of the American corporations was uneven. As their number increased, competition reduced their profitability. Risk, and therefore premiums, remained high throughout the Napoleonic Wars; between 1803 and 1812, as many as 1,600 American vessels were captured by various European powers.⁵⁹

After 1815, premiums fell to low peacetime rates, but fierce competition persisted and profits remained low until the 1840s, when the expansion of American shipping created a “golden period,” which lasted until the American Civil War.⁶⁰ Throughout, although there were apparently unsuccessful attempts to re-establish private underwriting, American marine insurance remained almost entirely in the hands of corporations.⁶¹ After the Civil War, as shipping technology shifted from wood and wind power to iron and steam, and the invention of the telegraph revolutionized information flows, the American shipping industry went into decline and the British companies and Lloyd's successfully re-invaded the American market; but that is another story.

HYPOTHESIS: A “LEMONS” PROBLEM

Given their advantages, why did the chartered corporations fail to dominate British marine insurance after 1720? Conversely, if private underwriting was somehow more efficient, then why did a Lloyd's-like institution for private underwriting not develop in the United States? After all, American underwriters and merchants were very familiar with Lloyd's.

One might hypothesize that the British corporations chartered in 1720 simply had an insufficient capital stock to capture the market. However, although their charters gave them the right to increase their capital stock to £1.5 million, neither ever did so.⁶²

The nominal capital of the Royal Exchange was only £508,329 during most of the eighteenth century, and the capital available for underwriting was even less.⁶³ Even in comparison with this capital stock, their underwriting was very modest. In 1761–1765, for example, annual premiums (also a rough measure of the expected value of losses) averaged only £36,600 at the REA, yielding average profits of about £6,100; the London Assurance averaged premiums of £43,300 and profits of £7,300.⁶⁴ Furthermore, both corporations were able to substantially increase their underwriting activity when the opportunity arose during the Napoleonic wars (Figure 1), and even then, were eager to extend their business further.⁶⁵ Finally, capital constraints would not explain the corporations' failure to drive Lloyd's out of business after 1824.

One factor contributing to the corporations' small share of the market, noted by both A. H. John and Barry Supple, is that underwriting was a “gaining trade,” so that it was to the mutual advantage of merchants to insure each other.⁶⁶

This can certainly explain some private underwriting, but would only apply to the minority of merchants who were also active brokers and underwriters (and as the American experience showed, the corporate form could equally be used as an institutional device for such risk-sharing).⁶⁷

In this section, I argue that the marine insurance market was characterized by multiple equilibria, and that the survival of private underwriters in Britain hinged on their superior access to the information needed to assess risks, which created a lemons problem for the comparatively less-well-informed underwriters at the corporations. In the next section, I argue that the timing of historical events influenced equilibrium selection, giving rise to a divergence in institutional structure between Britain and the United States.

The greater security and convenience offered by corporations was the key argument for the chartering of corporations in both Britain in 1720 and America in the 1790s. From merchants' correspondence and other contemporary records, it is clear that these were perceived as their main advantages, particularly in wartime. In 1748, one merchant wrote:

On a level playing field, therefore, corporations ought to drive private underwriters out of business. However, as emphasized previously, asymmetric information was a pervasive feature of marine insurance contracts in the eighteenth century (much more so than in the case of fire or life insurance). Information traveled slowly and was frequently incomplete, and merchants had strong incentives to conceal information that, if known, would raise the premium and to represent other facts so as to reduce the premium. The corporations' advantage in security might therefore be overturned if they were at a disadvantage in assessing risks.

In Britain, this is precisely what happened. As Lloyd's developed as a center for sharing information, the corporations found themselves increasingly at a disadvantage in evaluating risks. The business practices that evolved at Lloyd's—the reputation mechanism, the system of lead underwriters, the web of credit relationships—all tended to undermine the corporations' advantage in security while putting them at a disadvantage in evaluating risks.

To see how the lemons problem might develop, imagine a variety of merchants with differing degrees of private information about the voyages they propose to insure, and suppose that the private underwriters have access to some of this private information, while (for simplicity) the corporations observe none of the private information. Then the “best” risks—those merchants whose vessels are unusually seaworthy, for example—might prefer to insure with the private underwriters, despite the corporations' greater financial security, because the private underwriters, being better informed, are likely to charge them lower premiums. But then, the corporations will be left with a disproportionately poor selection of risks, forcing them to raise their premiums. This in turn will induce more of the better risks to apply to private underwriters; and so on, until the corporations are left with only the very worst risks. Thus, an equilibrium might develop in which the better risks are insured by private underwriters at low premiums, while the corporations charge high premiums and receive business only from the worst risks.

There is a further twist to this lemons problem. Basic insurance theory suggests that, at a given premium, good risks would wish to purchase a lower quantity of insurance than bad risks.⁶⁹

The historical record confirms that merchants frequently underinsured, particularly if they had a relatively small amount of merchandise traveling on a “good” ship.⁷⁰ Thus, not only will the less-informed underwriters (in this case, the corporations) get the worst risks, but the worse the risks are, the more insurance they will buy.

Of course, the information available to the private underwriters depends on the volume of business they transact. If we begin with the assumption that the private underwriters have no informational advantage, then given the insecurity of private underwriting, all merchants would indeed prefer to insure with the corporations, and the lemons problem would not develop. Thus, there is an alternative equilibrium in which all merchants choose corporate underwriters.

The lemons hypothesis can explain why the British corporations failed to dominate the British marine insurance market. Although they were not excluded from Lloyd's news service, the corporations could not match the collective expertise and local knowledge of the private underwriters.⁷¹

Neither could they write “lines” on a Lloyd's policy; they issued separate policies, with their own terms and conditions, according to their own rules. A bylaw implemented at Lloyd's in 1834 formalized the existing rule that all underwriting had to be strictly personal: “no subscriber shall underwrite risks for the account, benefit or advantage, either direct or indirect, of any public company.”⁷²

Consistent with the lemons hypothesis, as the eighteenth century wore on, the corporations became increasingly conservative and restrictive in their underwriting. This was manifested in several ways. On regular risks, their premiums were generally 20–30 percent higher than at Lloyd's. After experiencing numerous instances of fraud in the early eighteenth century, they became increasingly reluctant to insure cross risks, and refused to cover the risk of seizure in a foreign port—both risks for which a specialist knowledge of local circumstances was of vital importance, and the corporations disadvantage was therefore particularly acute.⁷³

They also limited the amount they would insure on a single voyage, and placed stringent restrictions on deviations from the planned voyage. For many merchants, it was this restrictiveness, even more than the high premiums, that induced them to insure at Lloyd's instead. One stated that Another merchant, having insured with the London Assurance Company, warned his correspondent that in case of accident, he should have the documents relating to the claim “well arranged authenticated and sent by two conveyances” because “these Gentlemen or rather this Company are rather particular in respect to vouchers either for an average or loss.”⁷⁵ They were too particular for one American merchant, resident in London, who explained to his partners in Maryland:

Both Supple and John also treat the corporations “lack of vigor” in underwriting as the major proximate cause of their modest growth, but neither explains this conservatism satisfactorily.⁷⁷

John hypothesizes that the private underwriters were “better able, and more willing, to accept all kinds of risks, simply because they could be spread over a large number of persons.”⁷⁸ Yet, the corporations had hundreds of shareholders, so if anything, they should have had an advantage in this regard. Supple suggests that the companies' caution may be attributable in part to restrictions on the freedom of action of their underwriter, though he does not explain the motivation for any such restrictions.⁷⁹

The key to understanding the corporations' cautious approach to underwriting is the lemons problem, which resulted from their disadvantage in evaluating risks. For example, in 1810, the chief clerk in the marine insurance department of the Royal Exchange Assurance Co., explained to a Parliamentary Select Committee that the company's practice was to underwrite a maximum of £5,000–10,000 on any one ship in the West India or Baltic fleets, because it was usually only offered the opportunity to insure about one-tenth of the ships in the fleet. If it could insure all the ships in the fleet, he claimed, it would “have no hesitation” in insuring £20,000–25,000 on each ship, “because it would be playing a more equal game.”⁸⁰

Furthermore, it was the existence of Lloyd's that created this lemons problem:

The corporations were also at a disadvantage in learning about the “character” of market participants. Although the Royal Exchange company would have liked to expand its business, it often refused to insure merchants whom it did not know well because their insurances were not “tendered fairly.”⁸²

Probably because of their disadvantage in evaluating the “character” of applicants for insurance, a good deal of the corporations' business was restricted to merchants and brokers who were also directors or shareholders in the corporation.⁸³

INSTITUTIONAL CHANGE AND EQUILIBRIUM SELECTION

Transactions cost economics suggests that market institutions have an advantage in governing transactions where information is dispersed and where transmitting information is costly. So, for example, “marine risks, with a wide variety of complicated risk factors generating a large premium, might be better handled individually,” as at Lloyd's.⁸⁴

However, as Oliver Westall notes, the transactions cost approach does not address the process of institutional change. In particular, it cannot explain why Britain and America, which were in close commercial contact and shared essentially the same technology and knowledge, developed different institutions to govern apparently similar marine insurance transactions.

The argument in the preceding section suggests the potential for multiple equilibria in the marine insurance market, which can account for both the success of marine insurance corporations in the United States after 1792, and the persistence of private underwriting in Britain after 1720 and 1824. But why did different equilibria emerge?

In both Britain and the United States, the corporations began brightly. In Britain, in the period before it obtained its charter (1718–1720), the company that became the Royal Exchange Assurance corporation was initially very successful, did a large amount of business, and was even accused of undercutting the market in order to drive private underwriters out of business.⁸⁵

However, “The deflation of the South Sea bubble quickly undermined the financial basis of the Royal Exchange Assurance,” and the value of its stock, which had risen from less than £20 per share in January 1720 to over £200 in August, fell below £10 by the end of the year.⁸⁶ Both of the newly chartered corporations had difficulty raising the capital needed to pay their debts to the King's Civil List (half of which was eventually written off), primarily because many of their subscribers, who had paid in only a fraction of the nominal value of their stock, were themselves in financial trouble; repeated efforts to attract new subscribers were largely unsuccessful. The London Assurance suffered heavy losses as a result of a storm that sank 12 Jamaica ships in October 1720, and was forced to borrow money to survive.⁸⁷ “… [T]he Companies' difficulty was the underwriters' opportunity,” and within a few years the corporations were doing only a small proportion of the total marine insurance business.⁸⁸

As discussed previously, the institutional structures and intelligence-gathering apparatus that enabled private underwriters at Lloyd's to deal with the information asymmetries and agency problems inherent in marine insurance took time to evolve. In particular, the quarter century of warfare from 1793–1815 provided a sustained stimulus leading to substantial institutional change at Lloyd's. By 1824 Lloyd's was well developed, and the new influx of corporations were unable to dislodge the existing equilibrium dominated by private underwriting.

In contrast, in America, the formation of joint-stock corporations became possible at a time when private underwriting was still at a comparatively early stage of development, and at a highly favorable time for marine underwriting. Because many of the most prominent merchants and brokers were instrumental in founding the corporations, they probably suffered little or no informational disadvantage relative to private underwriters, and therefore no lemons problem.⁸⁹

Like the British corporations before them, they began brightly and were accused of undercutting the market in a deliberate attempt to drive private underwriters out of business.⁹⁰ Unlike the British corporations, they succeeded.

Could a private underwriting institution such as Lloyd's have survived, once established, in the United States? One might argue that such a development would have been hindered by the lack of a “London” in America: America had several major ports and numerous minor ports, so that the marine insurance market was really a fragmented collection of local markets. However, merchants corresponded freely all along the east coast, engaged in joint ventures, and insured each other's voyages. New York merchant Gerard Beekman, for example, frequently brokered policies for Rhode Island merchants in New York in the 1750s, and in turn he sometimes insured his own voyages in Philadelphia, where he found the premiums lower than in New York.⁹¹

So, it is at least plausible that the several existing centers for private underwriting might have evolved into more developed institutions, given time.⁹²

As a major component of the nineteenth-century expansion of trade, the development of marine insurance is of interest in its own right. However, the development of the industry also illustrates the interplay between design and evolution in the creation of institutions and the potential for path-dependent institutional change.

Masahiko Aoki argues that institutional change frequently involves short, turbulent periods of deliberate institutional innovation and experimentation, interspersed with longer periods during which these experiments are gradually weeded out.⁹³

The development of the marine insurance business in the eighteenth century appears to fit this pattern.⁹⁴ Exogenous events provided an impetus for deliberate institutional innovation at particular points in time, such as the efforts to obtain charters for the British corporations during the South Sea Bubble in 1719–1720; the chartering of the American corporations in the 1790s in response to the new opportunities created by American Independence; and the development of a formal structure at Lloyd's in response to the growth of business during the Napoleonic wars. In many ways, however, it was the evolutionary changes, which occurred between these episodes of deliberate innovation, that ultimately shaped the direction of overall institutional change. In particular, the twin evolution of business practices at Lloyd's and British insurance law during the eighteenth century gradually generated relatively effective institutions, despite a lack of centralized direction or deliberate institution building, as merchants, underwriters, and coffee sellers gradually altered their perceptions and behavior in response to a gradually changing environment.