How accurate is Belloc's version of European economic history in The Servile State, with regards to the seizing of the monastic lands?

The main question I see which can be addressed here is did the monasteries control a third of the land within England at the time of Henry VIII?

The problem is that the answer to this is not directly resolvable through historical means. The historical records concerning the monasteries of the time are mainly gleaned from a taxation survey compiled for Henry in 1535, the Valor Ecclesiasticus. The problem is this work is mainly concerned with income, not property. Most of the estimations of the extent of the property holdings of the church at the time are calculations based on this income.

This type of calculation is extremely difficult in detail because every church property had different sources of income which reflected different forms of land use, so any extrapolation against the income as a whole is fraught with error. This means every individual looking at this question would likely develop a different solution.

A book from 1909, English Monasteries on the Eve of the Dissolution, Volume 1 By Aleksandr Nikolaevich Savin has some relevant information concerning some alternate attempts at an answer to this question. From page 87 (emphasis mine):

English authors have for the most part understood that estimates of monastic estates in acres meant very little to the English reader and therefore they try to convey an idea of the area of monastic lands by stating that it amounted to this or that fraction of English territory. All the fractions given by them all these tenths, thirds, and fourths do not really represent the area but only the relation of monastic income from land to the national income from land. The anonymous author of the Jacobite pamphlet believes that the monks were seised of seven tenths of England but this is only another way of saying that of the twenty millions of the national income the monks have fourteen millions. All his argument rests upon the calculation of the national income from land And daring though he is he does not venture to give what may be regarded as a trustworthy figure for the national income from land in the sixteenth century and it is just because of this that he translates the monastic income in the Valor Ecclesiasticus into the economic language of his own day. Nasmith pursues the same course Other English authors have compared the monastic income not with the income from land but with the whole of the national income. Lord Herbert had given utterance to his view that the monastic income amounted to one fourth or one third of the national income. Collier believes that even the estimate of one fifth is much too high he speaks of it as a tenth. Hume's supposition concerning the amount of the national income at the beginning of the sixteenth century, although utterly without foundation was for some reason largely accepted. Lingard although he dislikes Hume quotes this from him and the same is done by Cobbett and even Dixon. If it were at all possible to define the amount of the national income this method would be the more correct as the monastic income was not derived from land only. But we are utterly unable to discover what in 1535 was the total national income and all the above calculations are entirely without foundation

Another work, The Dissolution of the Monasteries as Illustrated by the Suppression of the Religious Houses of Staffordshire, By Francis Aidan Hibbert, relates much of the same confusion from various sources (emphasis mine), with various estimates of income or acreage:

The monastic income where figures have been given ranges from Speed's £171,300 to Burnet's £131,607. Abbot Gasquet says the monastic lands amounted to two million acres. A Jacobite pamphleteer of 1717 asserted that the monks possessed seven tenths of the whole land: more sober writers have estimated less extravagantly. JR Green said it was a fifth and Dr Gairdner says a third. Writers have often told of 'hordes of idle men and women' in the religious houses. Dr Gasquet affirms that the number was 8,081 with "more than ten times that number of people who were their dependents or otherwise obtained a living in their service:" the total population of England being some four millions this gives a proportion of one in forty three. It is obvious that there remains much to be desired in the way of definiteness and exactness on many points.

So even comparing the income of the monasteries to that of England as a whole is difficult, since we have various estimations of the national income as well. The footnote from Savin, (pp. 87-88) discussing Hume's calculations show his estimate for total income changed by 25% over various editions:

In the first edition of Hume's History (England under the house the Tudors, 1759, p. 222) three millions are mentioned; he says,

'the whole lands and possessions of England had, a little before this period, been rated at three millions a year; so that the revenues of the monasteries (Hume took from Speed the figure £161,100) did not really much exceed the twentieth part of the national income, a sum vastly inferior to what is commonly apprehended.'

But in the first posthumous edition of the 'History' (1778, iv. 182) four millions are given; four millions are given also in the 1854 edition (iii 245-6). The authors who repeat Hume's statement adhere for some unknown reason to three millions. Cobbett, Letter xvi p. 452. Lingard repeats Hume's figure but makes a sceptical remark (ii 98-100) Dixon, i 247 50 322. I do not understand Dr Cunningham's approval of Dixon's figures (Growth i 531) For the year 1535 Dixon simply took from Speed the figure £320,000 and from Hume £3,000,000 and divided the second by the first; in one passage he says that the monastic wealth is a half, and in another a third of the wealth of the whole Church. Th Rogers (A. and P. iv 113) supposes that the monks possessed one third of the national wealth.

But this just begins to touch on the issues. How do we derive a figure on land held from the income? Savin again looks at the difficulty encountered there (pp. 85-86):

If our authorities do not enable us to calculate exactly the number of acres belonging to the monasteries one would think that we might rely upon averages and the division. But how shall we determine the average income of an acre for all the monastic territory. Monasteries possessed salt works, mines. fisheries. and the income of these cannot easily be expressed in acres. Monasteries had considerable estates in towns where the income from a given unit of area was much greater than in the country this, can however be translated into acres though not without difficulty. To what extent does the income from town property increase the average income of monastic land? We can avoid the difficulty of course by eliminating the income from fisheries, mines, and town lands and by expressing in acres the agricultural income only, but how shall we calculate the average agricultural income. Professor Kovalevsky seeks to evade the difficulty by taking from Rogers the income of an acre of land under cultivation. But in the country the monks had not arable land only there were meadow lands pastures woods and waste and the various kinds of land differed considerably from each other in the income which they yielded If we divide the country income of the monasteries in order to find the average income from an acre of arable land we shall obtain not the actual area of monastic lands but only what would have been the area of monastic lands if they had all been arable Is it possible to calculate the average income of country land in general? We can ascertain the average income of arable land meadow wood garden and pasture in the sixteenth century But to pass from these to the average income of land in general further information is necessary we must know the average relation between the areas of the different kinds of land Let us imagine several monasteries with the same land area but with a different distribution of the land into arable pasture & c Let us suppose that every acre of arable land yields to all these monasteries the same amount of income Let us suppose that every acre of meadow brings in an income the same for all the monasteries but different from the income of the arable land let us make the same supposition with regard to the income of wood and pasture It is quite obvious that the average income of the land will differ in each monastery in spite of the fact that the income of each separate kind of land is supposed to be the same for all the monasteries The average income of the land will be highest in those cases where the amount of unprofitable land is least and it will be lowest where there is the highest percentage of unprofitable land Thus an inquirer must find out the average distribution of monastic land into the different kinds and here he will meet with an insurmountable difficulty arising from the authorities The Valor Ecclesiasticus gives hardly any material for deciding the question of the division of land into the different kinds and I am afraid that the information derived from all the other sources will also prove inadequate If the investigator be particular and strict difficulties must arise with regard to the authorities even in deciding simpler questions such as the average income of the different kinds of land It seems to me that even an approximate translation of a large income from land into acres is an almost hopeless task when dealing with the sixteenth century and I would add that as the monastic titles to the land were extremely variable and fractional the definition of the land area of monastic lands appears to be not only very difficult but also very unimportant.

We can see that attempting to estimate the land holdings as a derivative of the income is difficult ('almost hopeless') at best, due to variation in each monasteries holdings and how they were utilized.

So was 'in the past, one third of the capital (land) were held by the monasteries' an historically accurate statement? It would be a statement based of information which does not directly exist in the historical record. It fits in the range of estimates from various authorities, so is as 'correct' as any figure.

This is one of those numbers we just don't have enough information to know with any degree of accuracy. Extrapolating from what information we have is a valuable tool, I just hesitate to ever call it 'historically accurate'.