TWO GREAT COMMONWEALTH STATISTICIANS
The year 1999 marked the passing of two men, Keith Archer and Jack O'Neill, close friends and professional colleagues who had served in and later led the Commonwealth Bureau of Census and Statistics (forerunner of the Australian Bureau of Statistics) with great distinction, and shaped it into a modern national statistical office.
During the 1960s Archer and O'Neill were a formidable team, and it was during this period that the Bureau first earned its international reputation as one of the best statistical offices in the world. Archer was a natural leader, but many of the developments during this period (outlined below) would not have happened without O'Neill's strong support and his skills, in administration and organisation, to carry them through. O'Neill showed his own flair for leadership when he became Commonwealth Statistician after Archer's retirement.
Keith McRae Archer, Commonwealth Statistician from 1962 to 1970, died in Canberra on 1 April 1999. He was 93.
Archer was born in Launceston, Tasmania on 19 October 1905. Following the death of his father when Archer was only three, he was brought up by relations at Kimberley, Tasmania. After receiving his primary school education at the local school, Archer was awarded a scholarship to attend Launceston Grammar School. He successfully completed his secondary school studies and pursued his sporting interests, particularly in cricket and tennis.
On completion of his secondary schooling, in 1923 Archer started work in the Tasmanian Statist's Office, and part-time study at the University of Tasmania, where he completed a Diploma of Commerce. Shortly after Archer joined the Tasmanian Statist's Office, it became part of the Commonwealth Bureau of Census and Statistics as part of an integration arrangement between the Commonwealth and Tasmanian Governments. It was to be another 30 years before the other States agreed to similar arrangements. Tasmania was at the leading edge of statistical development in Australia and, as a consequence, was the 'home State' for a number of Commonwealth Statisticians.
After 10 years on a range of statistical work in Tasmania, Archer transferred to Canberra in 1933, on what he thought would be a short term basis, to participate in processing of the Population Census. He never returned to Tasmania to live, although he was a frequent visitor to his home State.
After working in a range of increasingly more senior positions, Archer was appointed Commonwealth Statistician in 1962 following the retirement of Sir Stanley Carver, a position he held until his retirement in 1970 because of ill health.
Under Archer's leadership, a number of important initiatives were launched which together transformed the Bureau into a modern national statistics organisation. Of these initiatives, four stand out.
Introduction of Computing
In 1963 the Bureau acquired the first major electronic computer installation in the Commonwealth bureaucracy, and one of the first in Australia. Archer fought hard for this, having seen the potential for electronic data processing through his long involvement (and some frustration) with punch card processing. The introduction of electronic computers led to incredible increases in productivity in statistical work. Archer's early influence was an important factor in the ABS continuing to be regarded as having one of the best information technology installations in Australia.
Integration of Economic Statistics
Archer gave strategic direction to the integration of official economic statistics. The statistics collections from businesses had been designed and managed as independent operations, without much effort to ensure consistency in data item definitions and classifications across the collections, and to avoid gaps and overlaps in their scope and coverage. Archer's leadership culminated in the conduct of the Bureau's first integrated economic censuses of businesses, in respect of 1968-69, covering a significant part of the Australian market economy. The ABS's economic statistics collections continue to be closely integrated, providing high quality source data for the Australian national accounts as well as many other users of economic statistics. Thanks to Archer's vision and leadership, the extent of integration of Australian official economic statistics is much greater than in most other developed countries.
International Recognition of the Australian Statistical Service
Under Archer's leadership, the Australian statistical service received increasing international recognition, and Australia began its significant influence on international statistical activities, including the development of international statistical standards and classifications-an influence it continues to exert to this day. Key events in Archer's time as Commonwealth Statistician were the election of Australia to the United Nations Statistical Commission and Archer's appointment as chair in 1968, also the year of Archer's appointment as chair of the Conference of Asian Statisticians. Australia hosted the 36th Session of the International Statistical Institute in Sydney in 1967; Archer chaired the organising committee for this important conference.
Statistics Cadetship Scheme
Archer's fourth major initiative, and perhaps his most important contribution to public administration in Australia, was the establishment of the statistics cadetship scheme. The scheme was established to attract highly competent young people to complete their university studies, at Bureau expense, prior to starting work with the Bureau. The scheme ensured a flow of highly qualified graduates into the Bureau, particularly in the 1960s and 1970s. Over time, many of these people were promoted both within the ABS and to senior positions in other agencies. A significant number rose to become departmental heads and deputy secretaries.
The cadetship scheme's impact on effective public administration has been profound and will remain a lasting legacy of Keith Archer's vision. The scheme was complemented by Archer's strong encouragement of Bureau staff to undertake tertiary studies to enhance their careers and the professionalism of the Bureau.
A stroke on Christmas Day 1969 cut short Archer's time as Commonwealth Statistician and led to his retirement in the following year. His service to Australia in statistics and public administration was recognised by an OBE in 1965 and a CBE in 1971. In 1969, he was made a Fellow of the Royal Statistical Society and in 1971 a Fellow of the Australian Computer Society.
Through hard work and determination Archer was able to recover much of his health. This led to his second career as Executive Director of the Australian Association of Permanent Building Societies. He represented Australia at the World Conference of International Building Societies in Rio de Janeiro in 1974 and San Francisco in 1979. On retirement from this position in 1979, he became Honorary Secretary of ACT Association of Permanent Building Societies.
Archer was a remarkable personality who left his stamp on the Commonwealth Bureau of the Census and Statistics, and its successor, the Australian Bureau of Statistics. He had enormous warmth and feeling for people, with a capacity to inspire great enthusiasm among Bureau staff for his initiatives. He had a great talent for remembering names and faces, and took a real interest in the activities of the families of staff members, even after retirement.
Archer 'managed by walking around' long before this became a management catchcry. Two or three times a year he would set out to visit every Bureau staff member at their workplace. These surprise visits often caught staff unaware, but the word soon spread that Archer was on one of his visits, which may have introduced some bias into his sample observations of the industriousness of Bureau staff.
Jack O'Neill, Commonwealth Statistician from 1972 to 1975, died in Canberra on 11 October 1998 following a long illness. He was 88.
O'Neill was born in Wynyard, Tasmania in 1910. He received his primary education at Wynyard State School and then boarded at St Virgil's College in Hobart for his secondary years. A fine student, he was also an outstanding athlete, tennis player, cricketer and footballer. Later he combined with Keith Archer to represent the University of Tasmania in inter-varsity tennis competitions. This was to be the beginning of a partnership that would last for many years in their working environment.
In 1927 O'Neill started work as a clerk in the then Commonwealth Bureau of Census and Statistics in Hobart. He also started part-time university studies in the Faculty of Commerce at the University of Tasmania, but ill-health prevented him from completing them, and put a temporary hold on his working career - he did not work for four and a half years while recuperating from his illness.
In 1937 O'Neill resumed his work with the Bureau, but in Canberra when Sir Roland Wilson was the Commonwealth Statistician. Although his whole working life was with the Bureau, O'Neill spent periods during the war years outposted as a statistical officer with the Bureau of Meteorology in Melbourne and the Food Control Unit of the Department of Commerce and Agriculture.
O'Neill worked in various statistical areas of the Commonwealth Bureau of Census and Statistics, and quickly rose to become Deputy Commonwealth Statistician. Keith Archer was then the Commonwealth Statistician.
Given his own considerable leadership qualities, it was natural that O'Neill should succeed Archer as Commonwealth Statistician following his retirement.
O'Neill inspired awe as well as affection from Bureau staff. Highly individualistic and somewhat larger than life, he was far from the stereotype of a public servant of his time. He showed great courage and determination in pursuing his career despite several setbacks because of illness. He could appear intimidating, particularly when waving his walking stick to reinforce a point, but he was also known for the kindness under his no-nonsense manner, and he had a sharp sense of humour. He was particularly supportive of young staff, and guided and regularly monitored their progress.
Legal structures for an official statistical agency
O'Neill had a fine instinct and judgement on management and statistical issues. He had an instinctive feel for the quality and credibility of the numbers underlying official statistics. He also had great political sense, and judgement about the most appropriate legal structures for an official statistical agency. This was of enormous value in one of his final roles as Commonwealth Statistician-oversighting the development of the Australian Bureau of Statistics Act 1975 to give effect to one of the key findings of the Committee on the Integration of Data Systems (the so-called Crisp Committee). This Act led to the creation of the ABS, and has become a model for the legislation underpinning many other official statistical agencies, most recently that of the Republic of South Africa.
Implementing the statistics cadetship scheme
As mentioned earlier, a key initiative in the development of the Bureau as a modern official statistical agency was the recruitment of young graduates and the encouragement of existing staff to pursue tertiary studies. O'Neill gave strong support to Archer to ensure the success of this innovative approach to securing and building the leaders of the future. Without O'Neill's superb administrative skills the statistics cadetship scheme would not have been such a success in developing a generation of leaders for the Bureau and the Public Service as a whole.
Support for research and development
O'Neill also gave strong support for research and development into new statistical methods, despite the opposition of those suspicious of new techniques. Seasonal adjustment of economic data may seem second nature now to users of time series, but in the 1960s there was widespread resistance, inside and outside the Bureau, to the introduction of these methods. O'Neill's determination, and instinctive support for methodological work, was the key factor behind the introduction of seasonal adjustment techniques in Australian official statistics. Earlier, he had been a great supporter of the introduction of sampling methods into the ABS, now a standard tool of trade of official statisticians, but not so in the 1950s.