Books / Digital Text

XV. Central Banking in the United States III: The National Banking System

1. THE CIVIL WAR AND THE NATIONAL BANKING SYSTEM

The Civil War wrought an even more momentous change in the nation’s banking system than had the War of 1812. The early years of the war were financed by printing paper money—greenbacks—and the massive printing of money by the Treasury led to a universal suspension of specie payments by the Treasury itself and by the nation’s banks, at the end of December 1861. For the next two decades, the United States was once again on a depreciating inconvertible fiat standard.

The money supply of the country totaled $745 million in 1860; by 1863, the money supply had zoomed to $1.44 billion, an increase of 92.5 percent in three years, or 30.8 percent per annum. The result of this large monetary inflation was a severe inflation of prices. Wholesale prices rose from 100 in 1860, to 211 at the end of the war, a rise of 110.9 percent, or 22.2 percent per year.

After the middle of 1863, the federal government stopped issuing the highly depreciated greenbacks, and began to issue large amounts of public debt. The accumulated deficit during the war totaled $2.61 billion, of which the printing of greenbacks financed only $430 million, almost all in the first half of the war.

The Civil War public debt brought into prominence in American finance one Jay Cooke, who became known as “The Tycoon.” The Ohio-born Cooke had joined the moderately sized Philadelphia investment banking firm of Clark & Dodge as a clerk at the age of 18. In a few years, Cooke worked himself up to the status of junior partner, and in 1857 he left the firm to go into canal and railroad promotion and other business ventures. Cooke probably would have remained in relative obscurity, except for the lucky fact that he and his brother Henry, editor of the leading Republican newspaper in Ohio, the Ohio State Journal, were good friends of U.S. Senator Salmon P. Chase. Chase, a veteran leader of the antislavery movement, had lost the Republican presidential nomination in 1860 to Abraham Lincoln. The Cookes then determined to feather their nest by lobbying to make Salmon Chase Secretary of the Treasury. After extensive lobbying by the Cookes, the Chase appointment was secured, after which Jay Cooke quickly set up his own investment banking house of Jay Cooke & Co.

Everything was in place; it now remained to seize the opportunity. As the Cookes' father wrote of Henry: “I took up my pen principally to say that H.D.’s (Henry) plan in getting Chase into the Cabinet and (John) Sherman into the Senate is accomplished, and that now is the time for making money, by honest contracts out of the government.”1

Now indeed was their time for making money, and Cooke lost no time in seizing the advantage. After wining and dining his old friend, Cooke was able to induce Chase to take an unprecedented step in the fall of 1862: granting the House of Cooke a monopoly on the underwriting of the public debt. Cooke promptly hurled himself into the task of persuading the mass of the public to buy U.S. government bonds. In doing so, Cooke perhaps invented the art of public relations and of mass propaganda; certainly he did so in the realm of selling bonds. As Edward Kirkland, author of Industry Comes of Age: Business Labor & Public Policy 1860–1897, writes:

With characteristic optimism, he [Cooke] flung himself into a bond crusade. He recruited a small army of 2,500 sub-agents among bankers, insurance men, and community leaders and kept them inspired and informed by mail and telegraph. He taught the American people to buy bonds, using lavish advertising in newspapers, broadsides, and posters. God, destiny, duty, courage, patriotism—all summoned “Farmers, Mechanics, and Capitalists” to invest in loans.2

Loans which of course they had to purchase from Jay Cooke.

And purchase the loans they did, for Cooke’s bond sales soon reached the enormous figure of $1 to $2 million a day. Approximately $2 billion in bonds were bought and underwritten by Jay Cooke during the war. Cooke lost his monopoly in 1864, under pressure of rival bankers; but a year later he was reappointed to that highly lucrative post, keeping it until the House of Cooke crashed in the Panic of 1873.

It is not surprising that Jay Cooke acquired enormous political influence in the Republican administrations of the Civil War and after. Hugh McCulloch, Secretary of the Treasury from 1865 to 1869, was a close friend of Cooke’s, and when McCulloch left office he became head of Cooke’s London office. The Cooke brothers were also good friends of General Grant, and so they wielded great influence during the Grant administration.

No sooner had Cooke secured the monopoly of government bond underwriting than he teamed up with his associates Secretary of the Treasury Chase and Ohio’s Senator John Sherman to drive through a measure destined to have far more fateful effects than greenbacks on the American monetary system: the National Banking Acts. National banking destroyed the previous decentralized and fairly successful state banking system, and substituted a new, centralized and far more inflationary banking system under the aegis of Washington and a handful of Wall Street banks. Whereas the greenbacks were finally eliminated by the resumption of specie payments in 1879, the effects of the national banking system are still with us. Not only was this system in place until 1913, but it paved the way for the Federal Reserve System by instituting a quasi-central banking type of monetary system. The “inner contradictions” of the national banking system impelled the U.S. either to go on to a frankly central bank or to scrap centralized banking altogether and go back to decentralized state banking. Given the inner dynamic of state intervention, coupled with the common adoption of a statist ideology after the turn of the twentieth century, the course the nation would take was unfortunately inevitable.

Chase and Sherman drove the new system through under cover of the war necessity: setting up national banks to purchase large amounts of U.S. government bonds. Patterned after the free banking system, the nation’s banks were tied in a symbiotic relationship with the federal government and the public debt. The Jacksonian independent treasury was de facto swept away, and the Treasury would now keep its deposits in a new series of “pets”: the national banks, chartered directly by the federal government. In this way, the Republican Party was able to use the wartime emergency, marked by the virtual disappearance of Democrats from Congress, to fulfill the long-standing Whig-Republican dream of a permanently centralized banking system, able to inflate the supply of money and credit in a uniform manner. Sherman conceded that a vital object of the national banking system was to eradicate the doctrine of state’s rights, and to nationalize American politics.

The Cooke-Chase connection with the new national banking system was simple but crucial. As Secretary of the Treasury, Salmon Chase wanted an assured market for the government bonds that were being issued heavily during the Civil War. And as the monopoly underwriter of U.S. government bonds for every year but one from 1862 to 1873, Jay Cooke was even more directly interested in an assured and expanding market for his bonds. What better method of obtaining such a market than creating an entirely new banking system, in which expansion was directly tied to the banks' purchase of government bonds—and all from Jay Cooke?

The Cooke brothers played a major role in driving the National Banking Act of 1863 through a reluctant Congress. The Democrats, devoted to hard money, opposed the legislation almost to a man. Only a narrow majority of Republicans could be induced to agree to the bill. After John Sherman’s decisive speech in the Senate in favor of the measure, Henry Cooke—now head of the Washington office of the House of Cooke—wrote jubilantly to his brother:

It will be a great triumph, Jay, and one to which we have contributed more than any other living men. The bank bill had been repudiated by the House, and was without a sponsor in the Senate, and was thus virtually dead and buried when I induced Sherman to take hold of it, and we went to work with the newspapers.3

Going to work with the newspapers meant something more than gentle persuasion for the Cooke brothers. For as monopoly underwriter of government bonds, Cooke was paying the newspapers large sums for advertising, and so the Cookes realized that they could induce the newspapers to grant them an enormous amount of free space “in which to set forth the merits of the new national banking system.” Such space meant not only publicity and articles, but more important, the fervent editorial support of most of the nation’s press. And so the press, virtually bought for the occasion, kept up a drumroll of propaganda for the new national banking system. As Cooke himself related: “For six weeks or more nearly all the newspapers in the country were filled with our editorials condemning the state bank system and explaining the great benefits to be derived from the national banking system now proposed.” And every day the indefatigable Cookes put on the desks of every Congressman the relevant editorials from newspapers in their respective districts.4

As established in the bank acts of 1863 and 1864, national banks could be chartered by the Comptroller of the Currency in Washington, D.C. The banks were free in the sense that anyone meeting the legal requirements could obtain a charter, but the requirements were severe. For one thing, the minimum capital requirements were so high—from $50,000 for rural banks to $200,000 in the bigger cities—that small national banks could not be established, particularly in the large cities.

The national banking system created three sets of national banks: central reserve city, which was then only New York; reserve city, for other cities with over 500,000 population; and country, which included all other national banks.

Central reserve city banks were required to keep 25 percent of their notes and deposits in reserve of vault cash of lawful money, which included gold, silver, and greenbacks. This provision incorporated the reserve requirement concept which had been a feature of the free banking system. Reserve city banks, on the other hand, were allowed to keep one-half of their required reserves in vault cash, while the other half could be kept as demand deposits in central reserve city banks. Finally, country banks only had to keep a minimum reserve ratio of 15 percent to their notes and deposits; and only 40 percent of these reserves had to be in the form of vault cash. The other 60 percent of the country banks reserves could be in the form of demand deposits either at the reserve city or central reserve city banks.

In short, the individualized structure of the pre-Civil War state banking system was replaced by an inverted pyramid of country banks expanding on top of reserve city banks, which in turn expanded on top of New York City banks. Before the Civil War, every bank had to keep its own specie reserves, and any pyramiding of notes and deposits on top of specie was severely limited by calls for redemption in specie by other, competing banks as well as by the general public. But now, all the national banks in the country would pyramid in two layers on top of the relatively small base of reserves in the New York banks. Furthermore, these reserves could consist of inflated greenbacks as well as specie.

The national banks were not compelled to keep part of their reserves as deposits in larger banks, but they tended to do so. They could then expand uniformly on top of the larger banks, and they enjoyed the advantages of having a line of credit with a larger “correspondent” bank as well as earning interest in demand deposits at their bank.5

Furthermore, in a way pioneered by the free banking system, every national bank’s expansion of notes was tied intimately to its ownership of U.S. government bonds. Every bank could issue notes only if it deposited an equivalent in U.S. securities as collateral with the U.S. Treasury. Hence national banks could only expand their notes to the extent that they purchased U.S. government bonds. This provision tied the national banking system closely to the federal government’s expansion of public debt. The federal government had an assured, built-in market for its debt, and the more the banks purchased that debt, the more the banking system could inflate.

The pyramiding process was spurred by several other provisions of the National Banking Act. Every national bank was compelled to redeem the obligations of every other national bank at par. This provision erased a severe free market limit on the circulation of inflated notes and deposits: depreciation increasing as one got farther away from the headquarters of the bank. And while the federal government could scarcely make the notes of a private bank legal tender, it conferred quasi-legal tender status on the national banks by agreeing to receive their notes and deposits at par for dues and taxes. And yet, despite these enormous advantages granted by the federal government, national bank notes fell below par with greenbacks in the crises of 1867, and a number of national banks failed that year.6

While national banks were required to redeem the notes and deposits of each other at par, the requirement was made more difficult to meet by the government’s continuing to make branch banking illegal. Branch banking would have provided a swift method for banks calling on each other for redemption in cash. But, perhaps as a way of blocking such redemption, interstate, and even more, intrastate, banking continued to be illegal. A bank was only required to redeem its own notes at its home office, making redemption still more difficult. Furthermore, the redemption of notes was crippled by the federal government’s imposing a maximum limit of $3 million a month by which national bank notes could be contracted.7 In addition, limits which had been imposed on the issue of national bank notes were removed in 1875, after several years of the banks' straining at the maximum legal limit.

Furthermore, in June 1874, the structure of the national banking system was changed. Congress, in an inflationist move after the Panic of 1873, eliminated all reserve requirements on notes, keeping them only on deposits. This action released over $20 million of lawful money from bank reserves and allowed a further pyramiding of demand liabilities. The result was a separation of notes from deposits, with notes tied rigidly to bank holdings of government debt, while demand deposits pyramided on top of reserve ratios in specie and greenbacks.

Reserve requirements are now considered a sound and precise way to limit bank credit expansion, but the precision can cut two ways. Just as government safety codes can decrease safety by setting a lower limit for safety measures and inducing private firms to reduce safety downward to that common level, so reserve requirements can serve as lowest common denominators for bank reserve ratios. Free competition, on the other hand, will generally result in banks voluntarily keeping higher reserve ratios. Banks now keep fully loaned up, expanding to the limit imposed by the legal reserve ratio. Reserve requirements are more an inflationary than a restrictive monetary device.

The national banking system was intended to replace the state banks completely, but many state banks refused to join as members, despite the special privileges accorded to the national banks. The reserve and capital requirements for state banks were more onerous, and national banks were prohibited from making loans on real estate. With the state banks refusing to come to heel voluntarily, Congress, in March 1865, completed the Civil War revolution of the banking system by placing a prohibitive 10 percent tax upon all state bank notes. The tax virtually outlawed all note issues by the state banks. From 1865 national banks had a legal monopoly to issue bank notes.

At first, the state banks contracted and withered under the shock, and it looked as if the United States would indeed have only national banks. The number of state banks fell from 1,466 in 1863 to 297 in 1866, and total notes and deposits in state banks fell from $733 million in 1863 to only $101 million in 1866. After several years, however, the state banks began expanding again, albeit in a role subordinated to the national banks. In order to survive, the state banks had to keep deposit accounts at national banks, from whom they could “buy” national bank notes in order to redeem their deposits. In short, the state banks now became the fourth layer of the national pyramid of money and credit, on top of the country and the other national banks. The reserves of the state banks were kept, in addition to vault cash, as demand deposits at national banks, from whom they could redeem in cash. The multilayered structure of bank inflation under the national banking system was now compounded.

Once the national banking system was in place, Jay Cooke plunged in with a will. He not only sold the national banks their required bonds, but he himself set up new national banks which would have to buy his government securities. His agents formed national banks in the smaller towns of the South and West. Furthermore, he set up his own national banks, the First National Bank of Philadelphia and the First National Bank of Washington, D.C.

But the national banking system was in great need of a powerful bank in New York City to serve as the base of the inflationary pyramid for the country and reserve city banks. Shortly after the start of the system, three national banks had been organized in New York, but none of them was large or prestigious enough to serve as the fulcrum of the new banking structure. Jay Cooke, however, was happy to step into the breach, and he quickly established the Fourth National Bank of New York, capitalized at an enormous $5 million. After the war, Cooke favored resumption of specie payments, but only if greenbacks could be replaced one-to-one by new national bank notes. In his unbounded enthusiasm for national bank notes and their dependence on the federal debt, Cooke, in 1865, published a pamphlet proclaiming that in less than 20 years national bank note circulation would total $1 billion.8 The title of Cooke’s pamphlet is revealing: How our National Debt May Be a National Blessing. The Debt is Public Wealth, Political Union, Protection of Industry, Secure Basis for National Currency.9

2. THE NATIONAL BANKING ERA AND THE ORIGINS OF THE FEDERAL RESERVE SYSTEM

After the Civil War, the number of banks and the total of national bank notes and deposits all expanded and, after 1870, state banks began to expand as deposit creating institutions pyramiding on top of the national banks. The number of national banks increased from 1,294 in 1865 to 1,968 in 1873, while the number of state banks rose from 349 to 1,330 in the same period.

As a summary of the national banking era, we can agree with John Klein that

The financial panics of 1873, 1884, 1893, and 1907 were in large part an outgrowth of ... reserve pyramiding and excessive deposit creation by reserve city and central reserve city banks. These panics were triggered by the currency drains that took place in periods of relative prosperity when banks were loaned up.10

The major effect of the Panic of 1873 was to cause bankruptcies among overinflated banks and in railroads that had ridden the tide of vast government subsidy and bank speculation. In particular, we may note the poetic justice meted out to the extraordinarily powerful Jay Cooke.

By the late 1860s, Cooke had acquired control of the new transcontinental Northern Pacific Railroad. Northern Pacific had received the biggest federal subsidy during the great railroad boom of the 1860s: a land grant of no less than 47 million acres.

Cooke sold Northern Pacific bonds as he had learned to sell government securities: hiring pamphleteers, for example, to write propaganda about the alleged Mediterranean climate of the American Northwest. Many leading government officials and politicians were on the Cooke/Northern Pacific payroll, including Rutherford B. Hayes, Vice President Schuyler Colfax, and the private secretary of President Grant, General Horace Porter.

In 1869, Jay Cooke expressed his monetary philosophy in keeping with the grandiose swath he was cutting in American economic and financial life. “Why,” he asked,

should this Grand and Glorious country be stunted and dwarfed—its activities chilled and its very life blood curdled by these miserable “hard coin” theories—the musty theories of a bygone age. These men who are urging on premature resumption know nothing of the great and growing west which would grow twice as fast if it was not cramped for the means ...

But four years later, the overbuilt Northern Pacific crumbled, and Cooke’s government bond operation collapsed. And so the mighty House of Cooke—“stunted and dwarfed” by the market economy—crashed and went bankrupt, igniting the Panic of 1873.11

In each of the banking panics after the Civil War, 1873, 1884, 1893, and 1907, there was a general suspension of specie payments. The Panic of 1907 proved to be the most financially acute of them all. The bankers, almost to a man, had long agitated for going further than the national banking system, to go forward frankly and openly, surmounting the inner contradictions of the quasi-centralized system, to a system of central banking.

The bankers found that the helpful cartelization of the national banking system was not sufficient. A central bank, they believed, was needed to provide a lender of last resort, a federal governmental Santa Claus who would always stand ready to bail out banks in trouble. Furthermore, a central bank was needed to provide elasticity of the money supply. A common complaint by bankers and by economists in the latter part of the national banking era was that the money supply was inelastic. In plain English, this meant that there was no governmental mechanism to assure a greater expansion of the money supply—especially during panics and depressions, when banks particularly wished to be bailed out and to avoid contraction. The national banking system was particularly inelastic, since its issue of notes was dependent on the banks' deposit of government bonds at the Treasury. Furthermore, by the end of the nineteenth century, government bonds generally sold on the market at 40 percent over par. This meant that $1,400 worth of gold reserves would have to be sold by the banks to purchase every $1,000 worth of bonds—preventing the banks from expanding their note issues during a recession.12

In addition to the chronic desire by the banks to be subsidized and cartelized more effectively, the large Wall Street banks, by the end of the nineteenth century, saw financial control of the nation slipping away. For the state banks and other non-national banks had begun to grow faster and outstrip the nationals. Thus, while most banks were national in the 1870s and 1880s, by 1896 non-national banks constituted 61 percent of the total number, and by 1913, 71 percent. By 1896, moreover, the non-national banks held 54 percent of the total banking resources of the country, and this proportion had grown to 57 percent by 1913. The inclusion of Chicago and St. Louis as central reserve cities after 1887 further diluted Wall Street’s power. With Wall Street losing control and no longer able to cope, it was time to turn to the United States government to do the centralizing and cartelizing with Wall Street exerting effective control of the monetary system through the power of Washington.13

In addition to the bankers, economists, and businessmen, politicians and political parties were all ripe for a shift to a central banking system. Economists participated in the general intellectual shift in the late nineteenth century from laissez-faire, hard money, and minimal government to the new concepts of statism and big government imbibed from Bismarck’s Germany. The new collectivist spirit became known as progressivism, an ideology also embraced by businessmen and politicians. Having failed to achieve monopoly positions on the free market, big businessmen, after 1900, turned to the states and especially to the federal government to do the subsidizing and cartelizing on their behalf. Not only that: The Democratic Party in 1896 lost its century-long status as the champion of laissez-faire and hard money. For statists and inflationists under William Jennings Bryan captured the Democratic Party at its 1896 presidential convention. With the disappearance of the Democratic Party as the libertarian party in American life, both parties soon fell under the statist, progressive spell. A new era was under way, with virtually no one left to oppose the juggernaut.14

The growing consensus among the bankers was to transform the American banking system by establishing a central bank. That bank would have an absolute monopoly of note issue and reserve requirements and would then insure a multilayered pyramiding on top of its notes. The Central Bank could bail out banks in trouble and inflate the currency in a smooth, controlled, and uniform manner throughout the nation.

Banking reform along these lines was considered as early as the beginning of the 1890s, and particularly favorable was the American Bankers Association and especially the larger banks. In 1900, President McKinley’s Secretary of the Treasury, Lyman J. Gage, suggested the creation of a central bank. Gage was formerly president of the American Bankers Association, and also former president of the First National Bank of Chicago, an organization close to the then-Rockefeller-controlled National City Bank of New York. In 1908, a special committee of the New York Chamber of Commerce, which included Frank A. Vanderlip, president of the National City Bank, called for a new central bank “similar to the Bank of Germany.” Similar recommendations were made the same year by a commission of big bankers set up by the American Bankers Association, and headed by A. Barton Hepburn, chairman of the board of the then-Morgan-controlled Chase National Bank.15

The Panic of 1907 galvanized the bankers into accelerating proposals for a new banking system. With intellectuals and politicians now sympathetic to a newly centralized statism, there was virtually no opposition to adopting the European system of central banking. The various shifts in plans and proposals reflected a jockeying for power among political and financial groups, eventually resolved in the Federal Reserve Act of 1913, which the Wilson administration pushed through Congress by a large majority.

Amid all the maneuvering for power, perhaps the most interesting event was a secret summit meeting at Jekyll island, Georgia in December 1910, at which top representatives of the pro-central banking forces met to hammer out an agreement on the essential features of the new plan. The conferees consisted of Senator Nelson W Aldrich (R., R.I.), a Rockefeller kinsman who had headed the pro-central banking studies of the Congressionally created National Monetary Commission; Frank A. Vanderlip of Rockefeller’s National City Bank; Paul M. Warburg, of the investment banking firm of Kuhn, Loeb & Co., who had emigrated from Germany to bring to the U.S. the blessings of central banking; Henry P. Davison, a partner of J.P. Morgan & Co.; and Charles Norton, of the Morgan-controlled First National Bank of New York. With such powerful interests as the Morgans, the Rockefellers, and Kuhn, Loeb in basic agreement on a new central bank, who could prevail against it?

One particularly ironic note is that two economists who played an especially important role in establishing the Federal Reserve System were highly conservative men who spent the rest of their lives attacking the Fed’s inflationary policies (though not, unfortunately, to the extent of repudiating their own roles in creating the Fed). These were University of Chicago professor J. Laurence Laughlin and his former student, then a professor at Washington & Lee University, H. Parker Willis. Laughlin and Willis played a large part, not only in the technical drafting of the bill and the Fed structure, but also as political propagandists for the new central bank.

  • 1. In Henrietta Larson, Jay Cooke, Private Banker (Cambridge, Mass.: Harvard University Press, 1936), p. 103.
  • 2. Edward C. Kirkland, Industry Comes of Age: Business, Labor & Public Policy, 1860–1897 (New York: Holt, Rinehart & Winston, 1961), pp. 20–21.
  • 3. Quoted in Robert P. Sharkey, Money, Class and Party: An Economic Study of Civil War and Reconstruction (Baltimore: Johns Hopkins Press, 1959), p. 245.
  • 4. See Bray Hammond, Sovereignty and an Empty Purse: Banks and Politics in the Civil War (Princeton, N.J.: Princeton University Press, 1970), pp. 289–90.
  • 5. Banks generally paid interest on demand deposits until the Federal Government outlawed the practice in 1934.
  • 6. See Smith, Rationale of Central Banking, p. 48.
  • 7. See ibid., p. 132.
  • 8. Actually, Cooke erred, and national bank notes never reached that total. Instead, it was demand deposits that expanded, and reached the billion dollar mark by 1879.
  • 9. See Sharkey, Money, Class and Party, p. 247.
  • 10. John J. Klein, Money and the Economy, 2nd ed. (New York: Harcourt, Brace and World, 1970), pp. 145–46.
  • 11. Irwin Unger, The Greenback Era: A Social and Political History of American Finance, 1865–1879 (Princeton, N.J.: Princeton University Press, 1964), pp. 46–47, 221.
  • 12. On agitation by bankers and others for the substitution of a central bank for the national banking system, see among others, Robert Craig West, Banking Reform and the Federal Reserve, 1863–1923 (Ithaca, N.Y.: Cornell University Press, 1977).
  • 13. See Gabriel Kolko, The Triumph of Conservatism: A Reinterpretation of American History, 1900–1916 (Glencoe, Ill.: The Free Press, 1963), p. 140.
  • 14. In addition to Kolko, Triumph of Conservatism, see James Weinstein, The Corporate Ideal in The Liberal State, 1900–1918 (Boston: Beacon Press, 1968). On the new collectivist intellectuals, see James Gilbert, Designing the Industrial State: The Intellectual Pursuit of Collectivism in America, 1880–1940 (Chicago: Quadrangle Books, 1972); and Frank Tariello, Jr., The Reconstruction of American Political Ideology, 1865–1917 (Charlottesville: University Press of Virginia, 1981). On the transformation of the American party system with the Bryan takeover in 1896, see Paul Kleppner, The Cross of Culture: A Social Analysis of Midwestern Politics, 1850–1900 (New York: The Free Press, 1970), and idem., “From Ethnoreligious Conflict to Social Harmony: Coalitional and Party Transformations in the 1890s,” in S. Lipset, ed., Emerging Coalitions in American Politics (San Francisco: Institute for Contemporary Studies, 1978), pp. 41–59.
  • 15. See Kolko, Triumph of Conservatism, pp. 146–53.
Shield icon library