Lesley Cook in 1958 wrote an account of the early history of Blue Circle that has become the standard narrative, recycled by many other writers. Underlying the account is the rapid evolution of industry production capacity in the first two decades of the 20th century, during which two concerted attempts to achieve a monopoly failed, overtaken by the more rapid expansion of competitors. The atmosphere of failure that hung over Blue Circle was not lifted until the Board coup d'état in the mid-1920s swept away the industry's "old guard".
Cook's account of the period drew heavily upon Henry Osborne O'Hagan's personal reminiscences, and on the more objective account by Macrosty. The capacity data in the account were very limited, being essentially restricted to the list compiled in Davis's second book (still in copyright), which was not yet published when Macrosty wrote his account. Other data consist of a few anecdotal snippets. Much of the argument hinges upon the interpretation of these data.
It must be noted that, apart from the "certified" production data, the figures are all estimates of unspecified precision.
Of the firms listed in the prospectus, three dropped out immediately after the launch (Note 3). Macrosty said "they represented a production of 140,000 tons on the Thames and Medway, and 50,000 tons on the Tyne".
In 1907, there was the national "Census of Production", which gave cement output as 2,886,000 tons (Note 4). In 1909, the second edition of A. C. Davis's Portland Cement was published. It included an appendix in which the cement manufacturers were listed, with their "approximated weekly tonnage outputs in 1907". The total of the listed data was 67,070 tons - or 3,353,500 tons for a 50-week year. The census production was only 86% of this, and it is clear that these "output" data were actually capacity. However, the Census of Production also implies that total capacity was at least 69,000 t/week, so both sources are questionable. The Davis List is as follows:
It will be noted that 79 companies are listed, although a number of these are allotted a zero capacity, and some (not all) of these had in fact ceased to exist. Some had more than one plant, and of the 35 plants included in APCM at its launch, 16 remained in commission, so the list implies a total of 99 plants (not all operational). The total of the tonnages (67,070) divides into three cohorts: APCM (24,000 - 35.8%), other capacity in London, Thamesside and the Medway (20,030 - 29.9%) and other capacity in the rest of the UK (23,040 - 34.3%).
Lesley Cook used these numbers to explain the position of APCM towards the end of its first decade. She says that APCM did not grow between 1900 and 1907. The calculation took the 1899 "certified" production of 1,404,569 tons , subtracted 190,000 tons (in fact, a capacity) for the three drop-out firms, leaving around 1,214,600 tons. This she compares with Davis's 1907 capacity of 24,000 tons per week. The stasis, she thought, was due to the fact that plant closures (19 out of 35) had cancelled the effects of expansions at the larger plants (Note 5). The total national production in 1899 she gets from the prospectus: Thames/Medway production was 1,700,000 tons and this was " upward of 80 per cent. of the entire output of Portland Cement in the United Kingdom", so the national total must have been 2,125,000 tons. She then concludes that APCM began with "over half" (in fact 57%) of the total, while in 1907 (using Davis's numbers) it had "only just over one-third" (in fact 36%). In the London area, she says (page 40), "outside companies had by 1907 a capacity of about 20,000 tons a week, which means that the APCM had only a little over half the capacity in the area in which they had hoped for a monopoly". The capacity of provincial competitors in 1907 (34%), she compares with "only about a fifth" in 1899.
She claimed that "not a great deal is known" about the expansion of competitor firms. A number of snap-shot accounts of the expansion of such firms is not balanced by any account of the parallel developments in APCM. The industry could be divided into five sectors:
The original purpose of the formation of APCM had been to create a monopoly that could control supply, set unchallenged high prices and thus generate high profits. It failed in this basic objective. At the same time, production technology advanced faster in the first decade of the 20th century than in any previous or subsequent decade. Kilns installed in 1900-1904 were already obsolete five years later. Firms had to continually invest in new capacity in order to keep up, but were reluctant to discard the old equipment that generated the cash for reinvestment. The result, in a sluggish pre-war market, was over-capacity and falling prices, so that few firms generated profits in excess of those needed for their investment programmes. APCM in particular made steady profits, but paid no dividends on ordinary shares. The result was a steady diminution of share prices, to as low as 6% of face value (Note 6).
In 1910, there still existed a belief that a monopoly could be achieved, and a takeover of the company's shares was orchestrated by the "69 Old Broad Street Group" controlled by John Wynford Philipps, J. S. Austen and M. B. Snell. This group's purchase of the company was based solely on their assessment of the under-valued share price. O'Hagan was of the entirely correct opinion that their interest was solely in "buying cheap and selling dear" without any regard to the details of the business, but nevertheless, money was made available for a second round of amalgamation.
The purchase of what became BPCM took place during 1910-1912. They must - surely - have realised that anything short of a complete industry buy-out would once again result in failure, but it was recognised almost from the outset that this would not be achieved. Once again, the significance of those firms left out was grossly underestimated. Again, Cook assesses the size of the new groupings in terms of Davis's 1907 data: APCM with 24,600 tons per week capacity (APCM had taken over Tolhurst), BPCM with 30,560, and the remaining 47 firms with 11,910. Thus, APCM and BPCM together held 82.2% of 1907 capacity. The obvious problem with this is that the 1907 data no longer represented the situation in 1911, although the amalgamators' intelligence may not have been much better.
It is interesting in retrospect to look at the actual data on capacity and output during this period. The cementkilns database contains data on clinker production and capacity. These differ slightly from cement data depending on the proportion of other components added to the cement, the only legitimate one being at this time gypsum (Note 7). Gypsum was only added at plants with rotary kilns, and at these, addition rates were less than 2% originally, rising to 2.5% by 1925, so the effect is small. The database data are in tonnes, and can be aggregated into the five sectors as above:
The annual data over a 30-year period are given in the appendix. The capacity and output varied as follows:
In capacity, APCM (A) continued to develop through the first decade, its relative size diminishing mainly because of the faster development of the firms (B, C and D) that subsequently made up BPCM. After the formation of BPCM in 1910-12, there was sluggish growth in the combine, while the independents (E) began at that time a very rapid growth which diluted the combine's share to around 60% - essentially its share for the next 70 years.
Actual production was characterised for APCM by a rapid growth in the period leading up to its formation, probably partly an artifact of the contract terms for the participants, followed by a period of tactical retrenchment. After 1912, the independents' expanded modern capacity achieved higher utilisation than the other sectors.
Installation of rotary kilns was the main disruptive change during the period. Rotary output was as follows:
The immediate impression is that uptake of rotary production was fairly uniform across the sectors, except that the independents continually lagged behind by two or three years. The "early adopters" incurred high costs associated with modifying or scrapping early inefficient designs, and any gains in production efficiency from the new technology were probably insufficient to offset these. While the early adopters ultimately gained a large step-change in efficiency, by 1912, the new efficient plant designs were available "off-the-peg" to the independents from suppliers who had a good understanding of plant operation, so they avoided the costly learning process. Outstanding examples of this were Aberthaw and West Thurrock.
It is interesting to re-examine the original assumptions discussed above in terms of the database information. Firstly, the claim that "upward of 80 per cent. of the entire output of Portland Cement in the United Kingdom is produced on the Thames and Medway". This underlay the proposition that control of the Thames/Medway industry was sufficient to control the whole national industry.
Not only was Thames/Medway capacity scarcely more than 70% of national capacity in 1899, but this was only a temporary peak associated with the amalgamation. In fact, during the previous three decades, the proportion varied little from 67%, and it only ever reached 80% briefly in the 1850s. It should be remembered that Portland cement originated in the North, and with William Aspdin's move to Gateshead, a large industry grew in the Tyne/Tees area, consistently averaging 10% of national capacity throughout the second half of the nineteenth century. The southern chauvinism that gave rise to a belief in the overwhelming importance of the Thames/Medway area was the first nail in the coffin of the APCM project. Promoters who rarely stirred from their City offices never challenged it. This silly attitude continued among Thamessiders throughout the twentieth century.
A second claim: "The undertakings which the Association acquires, and others with whom they will have working arrangements, are estimated to produce about 89 per cent of the total capacity of production of Cement on the Thames and Medway". This would appear to be sufficient to dominate the local market.
89% is an 8:1 dominance. 86% is only 6:1. The real data show that, even if all firms had joined as planned, the proportion achieved was significantly less than that promised, even at the time of the prospectus. All these missed targets conspired to ensure that APCM could not possibly have dealt the knock-out blow to the industry that was intended. Those firms that decided to stand on the sidelines probably understood this.
Note 1. O'Hagan saw the gathering of reliable data as an important part of his effort to bring about the amalgamation, since the companies on their own had failed to gain sufficient access to each other's data. How well he succeeded in this is debatable.
Note 2. Although many more were negotiated, the final firms with "working arrangements" were Lee's, Martin Earles, Queenborough and Wouldham.
Note 3. These were I. C. Johnson, Trechmann Weekes and West Kent.
Note 4. 1907 Census of Production, HMSO, pp 753-756 & 775-776. The census is not detailed, and notwithstanding its legal mandate, looks as though it might not have been comprehensive: cement industry companies were undoubtedly among those least likely to submit a truthful return. It gives cement production in 1907 as 2,886,000 tons, exports at 764,000 tons and imports 113,000 tons. The production was subdivided according to capacity, but capacity data were incomplete. Among those returns that stated the plant capacity, there were 1464 static kilns with a total capacity of 38,000 t/week, and 72 rotary kilns with a total capacity of 22,000 t/week. These two categories together produced 2,424,000 tons. Among those returns that gave no capacity data, there were 26 static kilns which made 72,000 tons, and an unspecified number of kilns of unspecified type which made 381,000 tons. The total of these production figures is 2,877,000 tons, which is different from the headline figure. Thus it is hard to make any precise statement of the makeup of the industry in 1907. Far from holding delinquent firms' feet to the fire, I suppose the census authorities thought themselves lucky that they got any data at all.
Note 5. The tendency to close smaller plants rather than develop them might be seen as the attempt of a monopoly to raise prices by restricting supply, but APCM also had the problem that, aside from the major upgrades at Swanscombe and Bevans, few of their plants had the raw material reserves needed for major upgrades. To the likes of O'Hagan, buying up "going concerns", the latter was not a major issue. Plant descriptions of the time often call raw material reserves "inexhaustible", but this assessment does not allow for an upgrade from, say, 200 t/week to 10,000 t/week, which would have been inconceivable to the average small operator in 1900.
Note 6. The price of £10 ordinary shares bottomed out at 12/6 in 1909.
Note 7. Chalk, limestone, sandstone and slag were routinely added in quantity, although this was rarely admitted, and for audit purposes was counted as clinker. The formula was "clinker used" = "cement made" minus "gypsum used". An amusing instance of this rule being imparted by a plant supplier to a beginner manufacturer is to be found in Leicester Archives.
|Text and notes for this article are in preparation.|
|Original content © Dylan Moore 2016: commenced 04/09/2016: last edit 15/10/16.|
NOTE: the database is still a work in progress. Although the broad picture presented is reasonably accurate, individual items of data are still liable to minor changes from time to time.
|Clinker Capacity tonnes|
|Clinker Output tonnes|