The expense of levying a land-tax, which varied with every variation of the rent, would, no doubt, be somewhat greater than that of levying one which was always rated according to a fixed valuation. Some additional expense would necessarily be incurred, both by the different register-offices which it would be proper to establish in the different districts of the country, and by the different valuations which might occasionally be made of the lands which the proprietor chose to occupy himself. The expense of all this, however, might be very moderate, and much below what is incurred in the levying of many other taxes, which afford a very inconsiderable revenue in comparison of what might easily be drawn from a tax of this kind.
The discouragement which a variable land-tax of this kind might give to the improvement of land, seems to be the most important objection which can be made to it. The landlord would certainly be less disposed to improve, when the sovereign, who contributed nothing to the expense, was to share in the profit of the improvement. Even this objection might, perhaps, be obviated, by allowing the landlord, before he began his improvement, to ascertain, in conjunction with the officers of revenue, the actual value of his lands, according to the equitable arbitration of a certain number of landlords and farmers in the neighbourhood, equally chosen by both parties: and by rating him, according to this valuation, for such a number of years as might be fully sufficient for his complete indemnification. To draw the attention of the sovereign towards the improvement of the land, from a regard to the increase of his own revenue, is one or the principal advantages proposed by this species of land-tax. The term, therefore, allowed, for the indemnification of the landlord, ought not to be a great deal longer than what was necessary for that purpose, lest the remoteness of the interest should discourage too much this attention. It had better, however, be somewhat too long, than