and supply favorable to shifting. In the second the average tax rate of each county has been subtracted from the various interest rates at which money was loaned in the county (for the year 1904, in which all the necessary data are available), on the hypothesis that, to the extent to which mortgage rates would be reduced below 4 per cent., to that extent investors would be forced either to shift the tax or, what is more probable, move their investments to other fields. The results according to the first test show that the excess of the mortgage rate over the tax rate was less than 4 per cent. in 36 of the 71 counties, containing 63.14 per cent. of the population of the State and representing 62.67 per cent. of the mortgage loans. According to the second test the net return to resident mortgagees, over and above taxes, would have been reduced below 4 per cent. in 62.96 per cent. of the cases.1

Of course, calculations of this kind must be taken cum grano; but, on the whole, I am inclined to believe that the last figures represent with some accuracy the extent to which double taxation of mortgages in Wisconsin would prove ineffectual or harmful.

The preceding comparison brings to view the greatest weakness of the double tax method as commonly employed. It levies upon credit instruments whose true value is shown on their face a rate of taxation maintained at an inordinately high level because of the under-assessment of other forms of property. This is not only inequitable, but it is largely responsible for such shifting of the tax as actually takes place. A mere glance at the tax rates collected in the various Wisconsin counties reveals how widely different is the quality of the assessment work and how

1 Measuring by the amount and not by the number of mortgages. It may be worth while to add that the net return would be reduced below 31 per cent. in the case of 50.53 per cent. of the loans, below 34 per cent. in 41.06 per cent. of the loans, below 3 per cent. in 34.90 per cent. of the loans, and below 3 per cent. in 26.08 per cent of the loans.

impossible it would be on the face to collect such rates of taxation without inviting disaster. In the year 19041 average tax rates varied from .6590 per cent. in Calumet County to 4.4268 per cent. in Iron County. The average mortgage rate for Iron County in the same year was only 6.65 per cent. Under such conditions extended argument about the probable effects of hypothetical mortgage taxation is superfluous. Levy such a rate upon mortgages, and either there will be no mortgages to assess or the tax will be passed on to the debtors.

It follows, then, that, if we are to have mortgage taxation at all, the rate of taxation should be strictly limited. Better even than this, of course, would be a provision for home rule by which such counties as desired to exempt mortgages could do so. Home rule being impossible, however, both expediency and equity unite in the demand that this one form of personal property, which must be taxed at full value to be taxed at all, be protected by limitation of the tax rate. A keen and able critic has suggested that in districts such as Iron County assessors would undervalue mortgages in the same degree as other property. Such an argument to my mind is equivalent to the admission that in such districts mortgages could not be assessed. It requires unceasing pressure under any circumstances to maintain the assessment of mortgages. Assessors do not like the trouble of consulting records, usually kept at only one place in the county, in order to unearth mortgages. They must be supplied with data by supervisors of assessment, State tax commissions, or other outside agencies, if they are to assess mortgages at all. This is particularly true in the less-developed districts, where tax rates are high and, money being needed to develop the country, mortgage taxes are particularly

I select that year because average rates happen to be computed. See the Wisconsin Report, pp. 77-78.

unpopular. To permit assessors, after being supplied with the facts, to scale values down which are plainly written in the bond would be suicidal. Upon this rock the double tax system of Minnesota foundered. Under the old Minnesota law, assessors were supplied by the registers of deeds with records of mortgages taxable in their respective districts. An experienced Minnesota official wrote to me: "The assessors generally pass upon the validity of the mortgage. If the mortgagee can convince the assessor that a certain mortgage has no value for taxation purposes, he passes it up, and it is not assessed. This is why the carefully framed provisions of Section 804, Revised Laws of 1905, have not succeeded in securing effective mortgage taxation in Minnesota." The only system of mortgage taxation which is at all justifiable is one in which the rate is fixed, and fixed low enough to leave small investors a net return on their loans at least equal to the interest rate paid by savings banks, trust companies, and that very restricted class of high-grade investments with which the small investor is familiar. If we must have mortgage taxation at all, the Pennsylvania system seems, on the whole, the best system now employed in any American commonwealth, altho the rate in Middle Western States, at least, might safely, in the opinion of the writer, be somewhat higher than four mills.

The Pennsylvania system, or, at least, the double tax method, with a low fixed rate, seems superior to the California method, or the recording tax, for the all-sufficient reason that it is less likely to bring about shifting. Under the California tax and under the recording taxes, practically every mortgage is taxed. By the double tax method, loans from foreign mortgagees and, usually, from banks, trust and insurance companies, both domestic and foreign, are untaxed. Tax every element of the supply, large amounts of which are almost perfectly mobile, and, accord

ing to every theory of incidence, the mortgage rate will be raised; while the more practical, concrete investigations of Professor Plehn1 and the New York Tax Reform Association indicate that the rise will probably be fully as great as the tax itself. Tax only one element of the supply, leaving free access to an infinitely larger source, and, by reason of the very mobility of capital upon which the preceding conclusion rests, interest rates cannot be seriously affected. There is not only unimpeded access to the national money market, but in Wisconsin at least there is unimpeded access (or was under the old law) to domestic banks, trust and insurance companies, and building and loan associations. How, under such circumstances, could the mortgage rate be raised 1.5 per cent. (the average rate of taxation throughout the State) by reason of taxation? It is true that the recording tax yields more, but it purchases productivity at the expense of the debtor, and, if the debtor foots the bill, the more productive the tax, the worse it becomes. It would be far better to exempt mortgages and secure the additional revenue by raising the general tax rate.

It is also true that the double tax method shields the debtor at the expense of the yield: it prevents the tax from becoming unbearable by making it partially ineffectual. But only partially ineffectual. There is a class of investors in mortgages-and in well-developed agricultural communities a very important class-who will pay a moderate tax rather than send their money outside of the State or change the form of their investment. They know the security of real estate in their own immediate neighborhood, and this is about the only class of securities of which they have any knowledge and in which they repose any real confidence. Frequently they make their

1 Yale Review, May, 1899.

2 Mortgage Taxation and Interest Rates, published by the New York Tax Reform Association, 1906.

loans when they sell their own land, taking a low interest rate, getting a high price, often mortgaging the property to a much higher proportion of its value than any outsider would consent to do, because "it used to be theirs, and they know what it's worth." Such loans, the so-called purchase money mortgages, must remain within the State; they constitute an integral part of the process by which land values are raised. The small investor of the class here described is not easily affected by minor variations of the interest rate. He wants his money to earn as much as possible, to be sure, but he is far more concerned about its security than its exact earning power. It is investors such as this, who own the three million dollars' worth of moneys and credits, mostly mortgages, which Clayton County, Iowa, keeps on its assessment rolls from year to year, who own, for the most part, the fifty-three million dollars' worth of mortgages which Michigan has on its tax rolls. It is this typical mortgagee whose existence and importance has been overlooked by those economists who confidently assume that any tax upon mortgages will inevitably be shifted. As a matter of fact, and particularly in agricultural communities, the supply of money seeking mortgage investment is far more persistent, far less mobile, far less elastic, than the average writer would have us believe. As a fiscal officer of great ability and long experience in one of the most important American commonwealths writes to me: "Some contend that the taxation of mortgages on a par with other property would result in an increase in the interest rate, but I do not think so. There is so much money seeking investments of that nature that I cannot believe that the interest rate would be affected thereby."

But, must we, after all, have any kind of mortgage taxation? Is it worth while to tax the small capitalist, who invests at home, and who has not the training to