THE INCIDENCE OF TAXATION IN THE UNITED

AT

KINGDOM.

T the present time various causes have combined to create a good deal of interest in the incidence of taxation in the United Kingdom. The question that has been and is being most keenly discussed is "what is a fair basis of taxation?" The results of the recent commission on the Financial Relations between England and Ireland caused many people to declare that Ireland was overburdened with taxation. Their opponents replied that a man with a given income in Ireland was taxed in just the same way as a man with the same income in England, and that, therefore, there could be no unfairness. But this answer is clearly not complete. What we have to show is that between people of different incomes the English system of taxation is substantially fair. To do this we should want a definition of "fairness." At present none exists; but that is not a sufficient reason for entirely ignoring the question, and it may be well to consider what the actual incidence of public burdens in the United Kingdom is, and then, having determined that incidence, to judge for ourselves whether it is fair. The question "what is the actual incidence?" is one that should be attacked by the economist; the more difficult one "what is fairness?" we are compelled to abandon to politicians. I propose, therefore, in this article to make an estimate of the incidence of public burdens, leaving, for the present at any rate, the more difficult inquiry.

In the last sentence I have used the expression public burden. By this I mean any tax, duty, rate or other exaction by the imperial or local authorities in the United Kingdom whose incidence is on the present inhabitants of the United Kingdom. I have been compelled to use this expression, which I confess I do not like, in order to avoid an apparent logical contradiction in what I am going to say. At first I intended to distort the word tax from its usual meaning so as to include both duties and rates. This would be objectionable in itself, but I found

that an even greater objection arose when, in the course of my argument, I contended that the land tax was not a tax. Assuming that I was right in my contention, I had a proposition which was formally self-contradictory, and must inevitably drive me on further into some sort of dialectic. To escape from this evil, I abandoned this misuse of the word tax, and resorted to the expression public burden.

A classification of public burdens is a necessary prelude to any investigation of their incidence. Such classification has already been made by the writers on public finance and taxation. For my purpose the distinctions drawn by the authorities between different kinds of public burdens are, in general, too detailed, but there are two important lines of division which are important for me. One great classification of public burdens is into imperial and local. This distinction may be

looked at from two points of view: We may consider where the money comes from-that is, what authorities raise it-or where the money goes to-that is, what authorities receive it. In what follows I have considered where the money comes from, and I therefore include as imperial all those monies which are raised by the imperial authorities and then paid over to the land taxation account. This distinction is simple and unimportant. The other distinction is important and difficult. If I pay a tax, rate, duty or any other public burden for which I get no benefit or only one in common with the rest of the community, the word "burden" is clearly appropriate. But suppose I pay for special services received from the state or local authority. Can we say that a municipal water rate is a burden? We do not consider a water rate paid to a private company as a burden; if the municipality acquires the water works, does the rate thereby become a burden? To give a complete answer to this we should have to consider what is the form of the economic organization of our country. Is it socialistic, or individualistic, or of some other and what kind? But to give a rough answer, we need only look at the question in a more superficial manner. A state or public authority can carry on a business. If that business was in the hands of a private individual or company there would be a certain average rate of

profit. The state can arrange that the receipts from its business shall be such that the profit of the undertaking shall be (1) negative, (2) zero, (3) positive and less than a fair rate of profit, (4) greater than a fair rate of profit. In different cases under different conditions it may be for the general good that one rather than any other of these alternatives is chosen. Whether in any particular case there is a public burden, a public bounty, or neither, is best considered in special cases, than as a mere deduction from general principles. One or two of these special discussions will follow later. As regards the definition of public burden, I say that when a public burden is not a burden it is not a public burden.

My predecessors in the task of estimating the incidence of the public burdens are many. Perhaps the most noteworthy estimates are those of Professor Leoni Levi and Mr. Dudley Baxter. Professor Levi has made estimates for the years 1842, 1862, 1882; Mr. Baxter's estimate is for the year 1867. In this paper the estimate is for the year ending March 31st, 1895. Since in some respects my methods differ from those of my predecessors, and since the value of any estimate must largely depend upon the method used, I propose, before I proceed any further, to sketch shortly what my method is in order to enable the reader the more easily to put his finger on my mistakes.

A complete, or even partial, discussion of the theory of the incidence of public burdens would, I think, be out of place, and at any rate it would tend to be a mere repetition of what is said in the text books. But, in order to avoid confusion, I must make a short statement of what, for the purposes of this paper, I have considered to be the money value of the burden of any public burden. In the first place a public burden costs something to collect. It may cost the government one million to collect eleven millions. Correctly speaking, the amount of the burden of the ten millions paid into the exchequer is eleven millions. I have assumed that it is ten millions. In the second place, even if a public burden cost nothing to collect, it is generally the case that the burden on the man who pays it is not fairly measured by the amount he pays. There are senses in which he may be less well off by more than the amount he pays.

In the third place, in the case of certain burdens, as import or export duties, they are or may be on those who are not members of the nation. Part, the whole or more than the whole of the 4d. per pound duty on tea imported into the United Kingdom may be paid by the foreign producer. I have assumed that the burden on the Englishman who buys tea is 4d. per every pound of tea bought minus the cost of collection of the duty. Finally, there are certain difficulties connected with time; in certain cases we may be paying for our ancestors or making our descendants pay for us. All these difficulties I neglect by estimating the total burden to be equal to the amount received by the imperial or local authority. Thus, if I find that £80,000,000 was paid into the public exchequer as proceeds of the government exactions for one year, I assume the burden of this to be £80,000,000 and merely apply myself to apportioning this burden between the different classes of the community. Of course, I have to consider very carefully what I shall take as my definition of a public burden and which of the different taxes, rates and duties fall within my definition; but once having done this, I start with a fixed sum of money which I have to apportion amongst the different classes of the community as best I can.

My next step is to define the classes into which I divide the community. My last step to determine the intermediate and final incidence of the burden upon each of these classes.

My candid admission that, at the outset, I have made many assumptions which are not merely dubious, but untrue, may perhaps dispose the reader of this article to pause and seriously consider whether it is worth his while to go on. In order to induce him to proceed I wish to point out that the question which I am most interested in answering is of one of the relative and not of the absolute burden of monies raised by the imperial and local authorities for public purposes, and that, therefore, it by no means follows that my proportions are wrong because my total quantities are.

To take an illustration: in many kinds of statistical inquiry we want to compare the average of one set of quantities with the average of another set. It is often impossible, or nearly so,

to determine in any given case theoretically what is the best kind of average, or practically what the exact value of that average is; but it does not much matter what kind of average we take, so long as we merely want to compare the two averages inter se. It is only necessary to take the same average in each case. So here, though I know well that I have wrongly estimated the total burden that falls on each class of the community, yet I feel some confidence that my estimate of the relative burdens (neglecting the difference in the value of money for different individuals) is tolerably correct.

In the first place, let us consider the imperial public burdens when one small point is explained. The point is this: certain items of receipt are appropriated for special purposes, and therefore do not appear in the finance accounts. Many of these items are small; those that are not, as the judicature fees, may in many cases be considered to be payment for services rendered rather than public burdens. At any rate, I have neglected these appropriations.

The first step is simple. What was the imperial income for the year 1894-95? The net receipts from the taxes, duties, undertakings and other items of income of the imperial authorities for that year were as follows:

but of these the following sums were due to the local taxation