A Firm Level Database of Irish Creameries,1897–1921

Abstract

We present a microlevel database of Irish cooperative creameries covering the period 1897–1921. The data were hand collected from the annual reports of the Irish Agricultural Organisation Society (IAOS) and contain information from 531 creameries and covering 49 variables. We perform some initial analysis of the data, finding considerable heterogeneity in the productivity of creameries as measured by the milk/butter ratio. We focus on differences between the four historical provinces of Ireland, finding that the south of Ireland (the historical centre of butter production) was on average less productive than the north at the start of the period, although this changes after 1913, when Ulster becomes the least productive province. These results present interesting avenues for future work, given the IAOS’ focus on founding creameries in the north of the island.

Introduction

Since the turn of the millennium there has been an increasing availability of digitised primary sources in historical research. ¹ This has coincided with an increased appetite for genealogy and local history, leading to greater availability of data on individual people. Academics have taken advantage of this detailed data and individuals’ records, leading to the publication of several innovative and important studies addressing major topics in Irish economic history such as intermarriage, long-run effects of famine, and emigration. ² However, the digitisation of business records, particularly firm level micro data, has not been undertaken to the same extent, and as a result the use of such micro firm data is still very much a novelty in quantitative economic and business history in Ireland, and elsewhere. For agribusiness, however, such micro data does exist for food processing firms at the turn of the twentieth century. Here we document a recently digitised database of statistics pertaining to cooperative creameries in Ireland from 1897–1921.

The Irish Agricultural Organisation Society (IAOS) played a large part in promoting and establishing cooperative creameries, based on the Danish cooperative model, throughout the island of Ireland, and the database we present below is taken from the annual reports of the Society, although it should be noted that not all cooperative creameries, and of course no proprietary creameries, were members of the IAOS. ³ In fact, conflict between the cooperatives and the propriety operators was to form a major part of the history of the expansion of dairying before Irish independence, when the government decided to enforce cooperation by buying out private firms, such as the large Cleeve's concern. ⁴ By the 1930s, however, top down interest in cooperatives was declining, the ideals of the movement were compromised, and cooperation was losing its distinctive ethos, becoming merely another type of business providing economic services to (particularly dairy) farmers. ⁵

A large database of Danish cooperative creameries has recently been compiled by Henriques et al., ⁶ and earlier work by Henriksen et al. and Lampe and Sharp has employed Stochastic Frontier Analysis to provide firm level productivity measures for smaller databases of individual creameries and farms. ⁷ More broadly, there has been an increased interest in the study of the cooperative movement both in Ireland and beyond. ⁸ Given the comparison often made between Danish and Irish dairying, ⁹ which often presents Irish dairying in a somewhat unfavourable light, we present here a novel database of Irish creameries from 1897–1921 which will, in the future, facilitate international comparisons; for example with other European countries, such as France, Finland, and the Netherlands. ¹⁰

The remainder of this paper proceeds as follows. In the next Section, we give a brief overview of dairy processing in Ireland, then in Section 3 we present a short history of the IAOS. In Section 4 we present an overview of the database and in Section 5 we present some descriptive statistics. Section 6 concludes.

Dairy Processing

The history of Irish dairying has been researched in depth in a number of previous studies, although the present work adds some important quantitative evidence to this. ¹¹ The province of Munster was the traditional heartland of the Irish dairy industry. At its core was the Golden Vale (or Vein), an area of rolling pastureland which was exceedingly good for dairy farming. ¹² Cork had long been the main centre of the butter trade and the Cork Butter Market (connected to the hinterland via butter roads) was located within easy access of Cork harbour. ¹³ Butter was traditionally produced in dairy farms via a tiring and labour-intensive process, and Irish butter was highly variable in quality between individual farms, but also depended on the season and on how long it took producers to fill a firkin (a type of barrel) that was used to transport butter to the market. ¹⁴ In the late nineteenth century the butter market in Cork lost its prominence due to lack of dynamism and a failure to adapt to changes in technology and organisation within the dairying sector. ¹⁵

These changes were the adoption of centrifugal separators and organisation in factory production. The steam-powered automatic cream separator enabled the centralisation of milk processing by small producers, and additionally meant that a more uniform product of a higher quality could be produced. The first commercial creamery in Ireland using the separator was founded in 1884, and was followed in 1889 by the first cooperative. ¹⁶ The transition from farm to factory production when it occurred was rapid; butter production increased from 2 million lbs in 1892, to over 20 million lbs in 1904, and by 1915 it had increased to 37 million lbs. ¹⁷ Much of this owed to a rapid rate of formation of private creameries, with their total number more than tripling to 84 in the years 1889 and 1890 and all still confined to the southern counties of Cork, Limerick and Tipperary, while the number of Poor Law Unions (PLUs) containing creameries rose from ten to twenty-one. ¹⁸

In part as a response to the loss of market share on the British market to Scandinavian producers, the Irish Agricultural Organisation Society (IAOS) was founded in Dublin in 1894, exactly five years after the first cooperative creamery was established in Dromcollogher, in the county of Limerick. ¹⁹ Initially, attempts had been made to establish cooperative stores along British lines. ²⁰ However, finding little scope for development of retail cooperatives, the IAOS focused on rural cooperatives and particularly cooperative creameries, whose rapid growth in Denmark impressed its founder, Horace Plunkett. ²¹ R.A. Anderson, the long standing IAOS secretary, claimed that the original promotors were, ‘quite unaware [that] the industrial remedy was already being applied in foreign countries’, but when Plunkett and Co. became aware of these efforts and research had been undertaken ‘they became far more confident of ultimate success’. ²² However, Irish farmers lacked organisational capacity and had become imbued by a culture of dependence on outside agencies, landlords, government officials or private operators such as merchants, shopkeepers or cattle dealers. ²³ The IAOS was a paternalistic top-down organisation, while this differed from cooperative formation in Denmark it was similar to cooperative development in France ²⁴ and Iceland ²⁵ as well as developing countries such as India after the Second World War. ²⁶ The Irish cooperative movement also had conflicting support from the catholic clergy, while generally supportive of productive cooperatives, such as creameries, ²⁷ there was greater hostility towards retail cooperatives which threatened traditional clerical patronage networks. ²⁸

Qualitatively, however, it must be acknowledged that dairy cooperatives in Denmark and Ireland operated in significantly different legal environments. Exclusive supply contracts (or vertical restraints), whereby members of cooperatives were contractually obliged to supply milk to cooperatives for a set period of time, were a key aspect of cooperative creameries in Denmark; ²⁹ effectively creating what would today be considered ‘cartels’. However, attempts to implement and enforce similar contracts in Ireland were fraught with complications, ³⁰ and were deemed to be a restraint of trade in a landmark ruling in the 1919 case of McEllistrim v Ballymacelligott Co-operative Agricultural & Dairy Society Limited. ³¹ In fact, Irish, or rather one specific cooperative in Kerry, dairy cooperatives were at the heart of UK competition law and set an important precedent that is surprisingly still foundational to modern competition law. ³² Another notable distinction between Denmark and Ireland relates to the liability of members in a cooperative, Irish cooperatives operated under limited liability (as in the title of the society in the above mentioned case) while their Danish counterparts operated under unlimited liability, leading to different incentive structures in both countries. Effectively exploring the influence of these qualitative differences requires comparison of the aggregates from comparable datasets.

The Irish Agricultural Organisation Society

The IAOS operated as an apex institution of the cooperative movement in Ireland, and it promoted the idea and ideals of cooperation through the organisation of local meetings and by enticing farmers to join. ³³ According to its founder, Horace Plunkett, the IAOS was expected not just to ‘support’ existing but also to ‘create’ ³⁴ cooperatives and even to ‘persuade’ ³⁵ people to adopt cooperation. The paternalistic approach of the IAOS can be seen in an early circular written by Horace Plunkett in October 1895 which stated that the IAOS:

promotes and organises Societies of Farmers on Co-operative lines throughout Ireland by sending organisers, free of charge, to address meetings locally convened, and otherwise explain what steps it is necessary to take. Farmers, unaided, cannot take these steps, nor can anyone who has not made a special study of Agricultural Co-operation explain to them the exact procedure which must be followed in order to organise a Society which will work harmoniously and be permanent. Moreover, for the first year or so or their working, the Committees of young Societies need some supervision and direction. Thus under our guidance, in Dairying districts, Societies of Farmers are formed to own and work for their own profit Creameries equipped with the costly machinery now essential for the profitable manufacture of butter, similar to those which have met with such success in Denmark and else where abroad. ³⁶

A decade later, a French commentator observed that the IAOS had created nothing except organising, advising and controlling, since the running of the creamery itself was at the behest of local farmers. ³⁷ Plunkett had originally hoped that the IAOS would be a short-lived institution and that ‘in five years at the longest – and we hope much before that – our [IAO] Society will no longer be required’. ³⁸ The hope had been for a federation of cooperatives but this never materialised.

In terms of locational choice, even before the IAOS was founded, both private and cooperative creameries had formed and located in the old butter heartland of Munster. Thus, early creameries, such as Dowdalls of Cork which had 13 creameries by 1888, were established by merchants associated with the traditional butter trade. ³⁹ The importance of Munster was recognised by Plunkett, who records in his memoirs that the strength of the dairy industry there was one of the factors which inspired his early efforts: “capitalists had seized the material advantages which the abundant supply of Irish milk afforded, and the green pastures of the Golden Vein were studded with snow white creameries which proclaimed the transfer of this great Irish industry from the tiller of the soil to the man of commerce”. ⁴⁰ A similar sentiment is echoed contemporaneously by R. A. Anderson, who stated ‘his industry is merely being developed in the interest of the entrepreneur proprietor, who competes with a Co-Operative Society’ because the capitalist investment ‘deprives the farmer of the power to control his industry’. ⁴¹

However, the IAOS’ job was made difficult by the fact that creameries with different but functionally equivalent organisational structures already existed in Munster. Also, making matters more difficult, butter traders, fearing for their privileged position, actively undermined cooperatives. ⁴² Unsurprisingly therefore, the IAOS quickly turned its focus away from the contested Munster market towards traditionally peripheral butter manufacturing districts. In 1895, the IAOS had opened one creamery in Ulster, in Glangevlin, County Cavan; the following year, there were six in operation in Ulster, although the initial Glangevlin creamery had already ceased trading. By 1897, however, the IAOS reported that the remarkable feature in the growth of the movement has been its extension in the north and north-west creameries which it was responsible for founding. ⁴³ Moreover, the IAOS was even involved in actively closing creameries established on marginal dairy land, since it believed that the available dairy land was reaching saturation point. ⁴⁴ Figure 1 illustrates the evolution of the cooperative movement in Ireland.

Figure 1.

Maps of Irish creameries in different years.

Besides the above issues, farmers themselves were also sceptical and they reportedly could not understand the fixation on butter production. Thus, at the 1897 AGM of the IAOS a delegate from Ennistymon, County Clare questioned this as there were some parts of the country where there was hardly any dairying. ⁴⁵ The delegate also considered it was a mistake to start a dairy society at Kilfenora in County Clare. The Society, he thought, would command more general support if they went into matters of that description, instead of giving so much importance to dairying. ⁴⁶ In fact, it was not until 1912 that farmers themselves began to solicit the IAOS to help them establish cooperatives, rather than the other way around. ⁴⁷

Despite these challenges, Plunkett saw the cooperative movement as an essential component of a wider Irish Revival, and that success was not possible without the ‘enlist[ing] on your side the new forces in Irish life’, namely what he referred to as the ‘Gaelic revival movement’, ⁴⁸ and the improvement of the material circumstances of Irish farmers through the power of cooperation. ⁴⁹ By forming cooperatives, Irish farmers would develop sufficient strength to escape the tyranny of the middlemen who set prices and controlled markets. ⁵⁰ In fact, the idea of self-help dominated the movement. As Plunkett articulated it: “The conviction has been more and more borne in upon the Irish mind that the most important part of the work of regenerating Ireland must necessarily be done by Irishmen in Ireland.” Cooperation also implied more general improvements such as the use of new technology, technical education, and generally a more efficient and ‘business- like’ method of agriculture. This was very important at a time when Irish exports were being battered by competition from cheaper and higher quality exports from Denmark and Holland. Plunkett's slogan was, ‘better farming, better business, better living’. By better farming and better business he meant more efficient methods, achieving market power through cooperation, but also better character (values of enterprise, fair dealings, working for the commonwealth). By better living he meant the building up of rural communities, the provision of educational and cultural opportunities and activities and pride in one's place (social capital). This approach effectively implied a revolutionary change in farming practice in Ireland as it now placed greater emphasis on business acumen. ⁵¹

Overview of the Database

As Irish dairying was heavily influenced by the activities IAOS, the standard source material for studying the dairy industry and cooperation is the annual reports of the IAOS. ⁵² Smith-Gordan and Staple refer to the IAOS as primarily a ‘propaganist body’ through publications and regular meetings, ‘and by all such forms of publicity as the funds will allow’. ⁵³ From the outset the IAOS's published reports furnished statistics from ‘the audited returns’ of Societies as they provided ‘some very valuable information’, ⁵⁴ and were particularly valuable as a marketing tool for the work of the IAOS. Plunkett laid out his vision for the IAOS as a compiler of information and an advisor to the farming community that was armed with ‘all the information which our Society has collected for the purpose of advising the Irish farmer how to combine with his neighbour for the improvement of his business’ and ‘provide useful aids to knowledge’ through various medium. ⁵⁵ He later observed that there was a demand from farmers for ‘details of its application to the particular branches of farming carried on in their several districts,’ and this required accountants to instruct cooperative societies ‘in simple and efficient systems of book-keeping and in the general principles of conducting business’. ⁵⁶

The IAOS was intended to be a purely developmental institution with a mantra that ‘the more business you introduce in politics and the less politics in business, the better for both.’ ⁵⁷ Plunkett's call for the IAOS to be a cross-community organisation meant that it had a ‘strictly economic’ focus ‘and no political or sectarian ends shall be promoted by word or deed’. ⁵⁸ The IAOS annual reports and its newspaper, The Irish Homestead, were the primary organ for this economic propaganda. The opening paragraphs of IAOS annual report set out the ‘progress’ of the IAOS and the ‘growth of the movement’ with specific reference to the statistics published in appendices of the report.

In an early article published in the Irish Homestead, Plunkett stated how the ‘publications of the [IAO]Society describe the success which the Irish Co-operative Diary Societies have already attained’. ⁵⁹ Figures within the cooperative movement regularly evoked the progress of the cooperative movement by frequent references to these statistics, which recorded information on the activities of cooperatives which were founded by and registered with the IAOS. ⁶⁰

We digitised data from these reports using the collection held in the National Library of Ireland. The data reported varied over time: in earlier and later years the data is rich with information on members, milk inputs, and butter outputs, but unfortunately this is absent between 1906 and 1912 when only a narrow range of statistics were reported. Then, from 1913–1920, the coverage again significantly improved and includes a wide array of statistics relating to inputs and output.

As mentioned above, these IAOS data exclude creameries that experienced alternative beginnings such as joint-stock companies operated by farmers, or cooperatives associated with rival institutions. This absence of data for non-cooperative producers is problematic but here we can draw on our previous work that attempted to assess the market share and production of other manufacturers. ⁶¹ By the early 1900s the majority of new creameries were cooperative and there is evidence of private companies transferring to the cooperative form. For example, O’Donovan notes how the IAOS was able to persuade joint stock companies that it was unwise to invest capital in creameries in Ireland leading to the sale of private creameries and conversion to cooperatives as early as 1901. ⁶² Admittedly, not all dairy companies heeded this advice and many of the large joint-stock creameries operated extensive branch networks, which were predominantly located in the south-west of the island in the Golden Vale. ⁶³

Early reports of the IAOS primarily provide descriptive information about the workings of the society and development of cooperatives in Ireland. The report provides an anonymised summary of dairy activity, but geographic information is not associated with it. ⁶⁴ IAOS annual report for 1898 is the first report to provide detailed statistics. ⁶⁵ The reported data is for the year 1897 and provides information on twenty attributes: name of cooperative, county of location, membership, paid-up share capital, loan capital including bank overdraft, reserve fund and accumulated profits, value of land and buildings after appreciation, sales of butter, sales of other goods, working expenses, net profit, net loss, number of gallons of milk supplied, number of pounds of butter produced, average produce per gallon (Ozs), quantity of milk to produce 1 lb of butter (Gals), average price per Gal. paid for milk (pence), average price per lb received for butter (Pence), percentage of working expenses as compared with trade, and finally cost of producing 1 lb of butter exclusive of value of milk (pence). The same information is reported for 1898, ⁶⁶ as well as for 1899 and 1900. ⁶⁷

Then, from 1901 there is a change in reporting convention and only information on name of society, county, membership, butter sales, and sundry sales are reported. However, in the following year's report, data for 1901 is reported as an appendix and contains the same level of information as IAOS. ⁶⁸ The 1904 and 1905 IAOS reports saw another break in both how the information is presented and what is recorded. ⁶⁹ The information presented is for the years 1902, 1903 and 1904 and is arranged by province and county, with three more attributes reported than in the previous years: date of establishment, number of suppliers in 1904, reserve fund and accumulated profits, and sales of agricultural requirements. There is a reduction in the information presented in IAOS reports for the years 1906 to 1910. ⁷⁰ These reports are arranged by county and province and attributes recorded are only the following: name of society, province, county, date of establishment, membership, paid-up capital, loan capital, and turnover.

Then, the reports begin to include more detailed information from 1911 onwards. The 1913 IAOS report contains information for 1911 ⁷¹ and the 1914 IAOS report contains for 1912. ⁷² The attributes recorded are: name of society, province, county, date of establishment, membership, paid- up capital, loan capital, butter turnover, other sales turnover, cream and milk turnover, affiliation fees, and subscriptions. The reports for the years 1913–1920 re- turn to the high level of attributes and recorded creamery statistics for 1913, 1914, 1915, 1916, and 1917 including the following attributes: name of society, date of establishment, number of shareholders, amount of share capital paid up, loan capital (including bank overdraft), butter turnover, other sales turnover, cream and milk turnover, net profit and loss, reserve fund, gallons of milk received, lbs of butter made, average produce (galls of milk to make 1 lb of butter), average prices for milk (gallon), average prices for butter (per lb), affiliation fees, and subscriptions. ⁷³

Putting together the dataset in a panel, including information for the creameries from 1897 to 1921, the data can be roughly divided into three categories:

Basic characteristics

Input/Output

Financial statistics

The data in production have been converted from gallons to kilos, for easier international comparison and interpretation. Table A.1. in Appendix A contains summary statistics for all the variables. Furthermore, the description of all variables included, as well as the years for which they are available, are in Appendix A (Table A.2.).

A Descriptive Analysis of the Productivity of the IAOS Creameries

As a first use of the data, we now turn to productivity, which was commonly measured at that time using the ratio of milk to butter, i.e., the amount of milk needed to produce a unit of butter. In order to investigate heterogeneity in productivity, which we consider on the provincial level, we use the following fixed-effects model:

$$MBrati{o}_{i,t}=\sum _j{\beta }_j[Provinc{e}_{it}=j]+\sum _t{\gamma }_t[Yea{r}_{it}=t]+{\epsilon }_{s,t},$$

(1)

where MBratio_i,t is the outcome of interest, the milk/butter ratio in creamery i and year t. The reference year is 1897, which is the first year, and Ulster is chosen to be the reference province, since it is the province in the northern part of Ireland with the highest intensity of cooperative creameries. If β coefficients vary significantly from zero, we provide evidence that variation within every year can be explained by regional dummies, i.e., there is significant variation in productivity among provinces.

In principle, the efficiency of butter production in terms of milk input is affected by two factors:

The potential of the creamery to extract as much butterfat as possible from milk, normally captured by the technology used in processing the milk.

The quality of the milk, i.e., the fat or cream content.

Our main specification does not capture possible heterogeneity induced from the differing quality of milk, which might be location specific; unfortunately, there is no available information regarding the breed of cows in the database. However, another determinant of productivity is the scale of production, captured by the number of members in the creamery. Therefore, the second specification takes the following semi-parametric form:

$$MBrati{o}_{i,t}=\sum _j{\beta }_j[Provinc{e}_{it}=j]+\sum _t{\gamma }_t[Yea{r}_{it}=t]+\delta logMembershi{p}_{it}+{\epsilon }_{s,t},$$

(2)

Finally, we might suspect that the heterogeneity among provinces changes over time, and this motivates the following flexible specification:

$$MBrati{o}_{i,t}=\sum _{j,t}{\beta }_{j,t}[Provinc{e}_{it}=jxYea{r}_{it}=t]+\delta logMembershi{p}_{it}+{\epsilon }_{s,t},$$

(3)

Table A.3. in Appendix A reports the results from the three specifications. Due to the gap in the data on production from 1904 to 1912, we have divided the results in two different periods, 1897–1903 and 1913–1921. Column (1) shows the results for specification (1) for the years 1897–1903, while Column (2) shows the result for specification (2) for the same years. Column (3) shows the results for specification (1) for the years 1913–1921, while Column (4) shows the result for specification (2) for the same years. Column (5) reports the results for the flexible approach, specification (3). The reference region is chosen to be Ulster, that historically had the largest intensity of cooperatives in the northern part of Ireland, and thus the other three regions are compared to this. There is considerable variation in productivity between provinces, as revealed by the strong and significant F- statistic. This does not change across specifications. It should be noted that although there are slight changes in parameter sizes when controlling for creamery size (specification 2), this does not explain away the qualitative conclusion: there is considerable regional heterogeneity in productivity. What is observed and should be highlighted is the dramatic change in the variation over the years.

In the earlier years (1897–1903), milk/butter ratios are larger in Leinster and Munster, indicating that the creameries in the south of Ireland were less efficient than the creameries in the north. Relative to Ulster, in Leinster around 1.1 kg of additional milk was needed to produce a kg of butter, while in Connacht around 0.7 extra kg is needed. When controlling for membership, the results look very similar and the pattern is the same. However, during the later years from 1913 to 1921, all three provinces are more efficient than Ulster. More specifically, in Leinster and Munster almost 0.5 kg less milk is needed to produce 1 kg of butter.

Specification (3) produces parameters for each province in each year. Figure 2 shows changes in milk/butter ratios attributed to each province compared to Ulster in 1897 as parameter estimates. The pattern that we discuss above is clearer in the graph. The provinces of Connacht, Leinster and Munster have a larger milk/butter ratio in the earlier years, while after 1913 they are smaller. This means that while creameries in Ulster are more efficient in the earlier years, creameries in the other provinces converge.

Figure 2.

Empirical results per province. Notes: The results are based on specification (5) of Table A.3. Province Ulster and Year 1897 are used as references. The dashed vertical line marks the break in the data between 1903 and 1913.

Conclusion

The provincial heterogeneity we uncover using the novel database requires some explanation. There were, of course, differences between the north and south, such as in terms of the availability of cows, average milk yields, infrastructure availability due to pre-existing butter production etc. ⁷⁴ One argument explaining the Ulster productivity advantage in the early years is that the cooperative organisational form, predominant in Ulster, led to greater productivity. ⁷⁵ However, this argument relies on aggregate evidence that does not account for the sizeable condensed milk production located in Munster. What may appear a more realistic explanation is the smaller concentration of creameries in Ulster which enabled closer monitoring of milk supplies. ⁷⁶ Competition between creameries in the south led to the acceptance of milk of lower quality. ⁷⁷ The introduction of binding rules may have helped to improve the quality of milk received in the south of the island, although these rules were later changed. Furthermore, what is interesting is that the IAOS’ focus on Ulster seems to have backfired, which runs contrary to a view that Ulster was the most innovative region in terms of cooperation. ⁷⁸ Perhaps too many creameries were founded in that province, reflecting some overenthusiasm of the IAOS? Or perhaps Ulster's natural geographic disadvantages became more apparent over time? Whatever the case, data is now available to provide much more detailed studies of this and many other topics.

While extending the database beyond 1920 would be ideal, it is worth bearing in mind how the dairy industry changed in north and south in the 1920s and onwards. In the south, it became dominated by cooperatives, which is somewhat surprising given the strength of proprietary creameries there. A recent history of the Dairy Disposal Board, established to nationalise and rationalise the Irish dairy industry in the Free State in the 1920s, explains this transition along normative lines as being a consequence of the ‘milk wars’ between cooperatives and proprietary creameries in the 1900s and 1910s. Ó Fathartagh argues that ‘unlike in Denmark…proprietary creameries were generally in competition with what co-operative creameries there were. Curtailing the proprietary creameries would be a major prerequisite to rescuing the Irish dairy industry.’ ⁷⁹ However, this view implicitly assumes that the cooperative organisational form was superior to the proprietary form. Yet, the findings of Henriksen et al. ⁸⁰ and the arguments outlined heretofore in this article place this assumption in considerable doubt. Thus, post-independence, the situation changed dramatically. O’Rourke ⁸¹ cites the overwhelmingly cooperative nature of Irish dairying post-1920 as evidence that Ireland, once independent, had no barriers to cooperation as there was elimination of rural/sectarian conflict. ⁸² However, this had in fact much to do with the state regulation of the dairy industry post-independence. With the establishment of the Irish Free State in 1921, state policy aimed deliberately to ‘make the manufacturing of butter entirely co-operative and place the ownership of the premises and direction of the industry in the hands of the farmers who supplied the raw material’. ⁸³ By 1926, out of 580 creamery and separating stations, 400 were cooperatively owned. ⁸⁴ Then, in 1927 the government founded the semi-state Dairy Disposal Company to rationalise the industry. The IAOS contributed directly to legislative effort as the then IAOS secretary, Henry Kennedy, was brother-in-law of the then Minister for Agriculture, Patrick Hogan. ⁸⁵ Cleeve's and other smaller concerns were effectively nationalised under this new body, until the 1970s when the Dairy Disposal Company was broken up, and ownership was transferred to a number of farmer cooperatives. The database provided here gives an opportunity to explore the outcome of the policy by comparing the performance of cooperative dairies north and south of the border pre- and post-rationalisation and nationalisation.