by a three-fourths vote. The Buffalo charter allows overdrafts in the following terms:

"The expenditures for each department, office, or other purpose during the fiscal year, shall be kept within the estimate made for it, except that in cases where the mayor, comptroller, and treasurer shall certify in writing that a greater amount than provided for in the estimates is necessary in any department of the city, the expenditures in any such department may be increased by the amount so certified by a two-thirds vote of the members elected to each board composing the common council, which vote shall be taken by calling the yeas and nays, and shall be entered upon the journals of the common council."

The most prevalent means of correcting the budget is by transferring money from one appropriation to another. This avoids the irregularity of the overdraft, and at the same time uses any actual surplus in the treasury to fill the deficit in certain departments. The general rule in New York, Wisconsin, and Iowa is to prohibit transfers altogether; Michigan and Minnesota prohibit them for some cities. Boston allows transfers on the written recommendation of the mayor approved by a twothirds vote of the council; but the budget order always provides for transfers by the auditor and mayor between the items of any department and, during the last two months of the year, between departments. Richmond likewise requires a two-thirds vote. In Ft. Wayne the council may make transfers on the recommendation of the comptroller. The practice is often carried to great excess; at Saginaw, for example, out of forty-eight funds, transfers were made in 1897-98 from twenty-seven funds, and transfers to eighteen funds; eight funds both received and gave transfers. The amount transferred was $56,454.78 out of a total expenditure of $712,712.80.

Certain funds are nearly always exempt from transfers, such as money realized from the sale of bonds, the receipts of productive enterprises like waterworks, and money raised for schools and for departments under state commissions.

No vital objection can be made to transfers if they are kept within narrow limits. The best way to limit them is probably to require a recommendation from the comptroller or the mayor or both, and a two-thirds or three-fourths vote in the

council. But transfers do most emphatically complicate the accounts and help to render the published report a mystery to the uninitiated.

By all odds the most direct and the least confusing mode of giving elasticity to the budget is to leave an adequate margin of revenue which can be used for exigencies in any department. This mode is adopted in Boston. The appropriation bill provides for a "reserved fund" amounting to from a fourth to a half of one per cent of the total appropriation; "and the city auditor is hereby authorized to transfer from this fund for current expenses only, as the mayor may direct, with the approval of the Committee on Finance." The law governing the city of Cincinnati requires that in the semiannual budget $50,000 shall be set aside as a contingent fund to meet deficiencies; this can be used only by a two-thirds vote of the council and with the approval of the mayor. In Minneapolis twenty votes out of twenty-six are necessary to make an appropriation from the contingent fund.

One small disadvantage of a reserved fund like this is that it requires the city to keep on hand a somewhat larger cash balance than when any surpluses that exist can be made use of by transfers or by the overdrafts of other departments. But the serious difficulties with such a fund are the political ones. When the makers of the budget are struggling to squeeze the estimates of the departments within the limits of the city's revenue, they must have great self-control to leave a dollar of possible revenue unappropriated; a city council that will do it voluntarily is rare indeed. On the other hand, when the charter requires that certain miscellaneous revenues shall go into such a reserved fund, the fund may become needlessly large and offer an irresistible temptation to spoliation. Newark affords a curious illustration of this. The contingent fund is fed by the license fees and a few other sources of revenue. It is very large, amounting to nearly ten per cent of the entire income of the city. When the council passes the tax ordinance, it also passes a "supplementary" resolution distributing the greater part of this fund to the various departments. Then even when the budget makers are able to leave a small sum unappropriated, the officers who control expenditures will have no peace until it is gone; as

long as it stands on the accounts unused it invites onslaughts by advocates of innumerable projects - absurd, visionary, partisan, or mercenary, and some really meritorious but all beyond the city's ability to pay for. Most officers who have had experience with such matters are unwilling to stand the pressure and prefer to have no reserved fund, notwithstanding the incontrovertible reasons for having one. The writer has found that nearly every other official who performs the part of "watchdog of the treasury" has some secret or roundabout means for accomplishing the purpose of a reserved fund, such as to intentionally underestimate the miscellaneous revenues or overestimate some item of expenditure, or keep still about some surplus which he knows will not be needed, but which is ostensibly pledged. He can then exhibit the accounts to those clamoring for money and challenge them to show where the money can come from, while he knows all the time where it might come from. Here also may be a reason why some comptrollers do not wish the accounts to be too simple or easily understood; it would not then be so easy to plead poverty to the importunate alderman or department chief.

So far only additions to particular portions of the budget have been considered. Transfers cause no change in the aggregate; supplementary appropriations out of surplus resources do not cause a deficit; the chief end of the budget to keep expenditure within the income-is attained. But a leak soon becomes a torrent, and if one department succeeds in spending more than its allowance, others will strive to do the same. It is by additions to the estimated expenditures, or by a shortage in the realized income as compared with the estimated income, that the city is made to close the fiscal year with a deficit.

There are several ways of allowing the total outgo of the year to exceed the total resources. The simplest one is to merely let the bills remain unpaid till the next fiscal year. This is a trick easy to accomplish and difficult to detect while in progA device, formerly much used but now happily disappearing, is to issue the usual orders or warrants to the claimants, but let them remain outstanding till some money comes in. . . . The most businesslike way is to issue certificates of indebtedness

ress.

or make loans at the banks, such as are frequently used to anticipate the taxes during the year, and let them run over into the next year. In some of the Southern and Western states the law allows expenditure to be made in excess of the amount specified in the budget, provided it is sanctioned by the voters at a special or general election. A law to that effect was recently passed in South Carolina. But the end is always the same: a floating debt is accumulated and future generations are required to pay for the folly or extravagance or corruption of the past. A floating debt, once started, tends to grow; some unpaid bills, outstanding warrants, or certificates of indebtedness must go over into the next fiscal year anyway; they thus show the way to making the expenditures of the year exceed the revenues.

CHAPTER XXVI

THE CUSTODY AND DISBURSEMENT OF PUBLIC MONEY

88. The Methods of the Federal Government. - Professor J. B. Phillips, after a careful study of the various methods employed by the federal government, presents the following conclusions: 1

Having now described the various methods of keeping the public money which have been employed since the organization of the government, it is proper to conclude this historical survey with some discussion of the relative merits and demerits of each. In order that such a discussion may be as effective as possible, perhaps it will be best, first, to set forth as clearly as may be, the chief requirements of an effective plan for the administration of the public funds. With these requirements in mind, it will be easy to test each of the several methods by this standard, and ascertain its advantages and defects. The chief requisites in a perfect system of keeping the public money appear to be the following:

1. Whatever the system employed, it should provide for the absolute safety of the public money. This is so obvious that it needs no comment. However great the other advantages of a system, they cannot compensate for a defect which casts the slightest suspicion on the safety of the funds.

2. The money should always be under the absolute control of the government. By this is meant that the drafts of the government will always be promptly met. No system which does not admit of this should be considered for a moment. The great problem is to provide such a system as will insure prompt payment, and at the same time cause no friction in the money market.

1 Methods of Keeping the Public Money of the United States, 138 et seq. In Publications of the Michigan Political Science Association, Vol. IV. Reprinted with consent of the author and the Association.