UK-PMSG (UK Productivity Measurement Study Group)

Editor:Ali Emrouznejad


  • A New Study Group of the UK Operational Research Society:

    Almost all organisations now either employ themselves or are the subject of some sort of performance measurement scheme.

    Operational Researchers have played a large and increasing role in such activity, applying methods such as expert systems, neural networks, data envelopment analysis and a variety of parametric statistical techniques. Applications can be found in almost any segment of the economy, but have been particularly notable in the public sector, health care, the regulated utilities and financial services.

    In the light of these developments, the UK Operational Research Society has recently established a Productivity Measurement Study Group. The aim of the Study Group is to disseminate advances in all forms of productivity measurement to as wide an audience as possible.

    *The primary focus will be on practical applications rather than theoretical developments.*

    A committee has been set up to co-ordinate the activities of the Study Group. It consists of Emmanuel Thanassoulis (Warwick University), Peter C. Smith, (York University (currently visiting St Andrews University)), Rod Green (Bath University) and Bruce Hollingsworth (Newcastle University).

    The Committee is pleased to announce that Professor Maurice Shutler, formerly Chief Industrial Advisor to the Monopolies and Mergers Commission in the UK has agreed to give the inaugural talk of the Study Group. Details of the talk appear at the end of this announcement.

    A programme of further meetings is now being put in place. Titles, speakers dates and venues will be announced in the OR Newsletter and elsewhere as soon as details have been finalised. The committee would welcome suggestions for future events, and is keen to establish a network of people who would like to be kept informed of the activities of the Study Group, both within and outside the OR Society.

    Anyone interested should contact the Study Group Chair, Dr Emmanuel Thanassoulis, Warwick Business School, University of Warwick, Coventry, CV4 7AL, UK, phone +44 (0)1203 522145, FAX +44 (0) 1203 524 539, e-mail ORSET@WBS.WARWICK.AC.UK.


  • Meeting 16 December 1998:

    1998 Meeting 16 December 1998 at 5.00 - 6.30pm Room A144 (Main Building) London School of Economics

    This presentation will be given by Professor H. David Sherman, Northeastern University, Boston, USA.

    "Improving Profitability of Bank Branch Networks: Benchmarking Productivity, Quality, Marketing, and Profitability Using Multistage Data Envelopment Analysis (DEA)". H. David Sherman , College of Business Administration, Northeastern University, Boston, MA, 02115, U.S. Tel. 617-373-4640, Fax 617-558-1235,

    Email: hsherman@lynx.neu.edu

    Abstract:

    Bank branch networks have significantly increased profitability using DEA by helping managers identify and implement heretofore unseen methods to improve branch operations. Via applications to networks ranging from 50 to 1500 branches, this powerful benchmarking technique evolved from an initial focus on productivity to incorporate quality, marketing effectiveness and profitability.

    The presentation will describe a bank branch application and address the following questions:

    1. How DEA can be applied to a branch network?

    2. How is the multi-stage analysis structured?

    3. How are the DEA results interpreted and communicated to management?

    4. How are the results translated into real operating improvements and increased profits?

    5. What are the current limitations from a technical, behavioural, and political perspective?

    6. Are bank branch experiences with DEA meaningful for other industry applications?

    7. Why isn't DEA used more in service businesses?

    The future developments are likely to generate new more challenging global best practice benchmarks incorporating productivity, service quality, marketing effectiveness, and profitability. This may help service organizations leap beyond current best practice performance levels. This evolution to a more comprehensive benchmark of performance, referred to as multidimensional balanced benchmarking (MBB), will be introduced with examples of the insights from one branch application.

    Attendance of the seminar is free.


  • Meeting in October 1997:

    The meeting of the Study Group will be addressed by

    16 October 1997 at 5.15 - 6.30pm, Room L4.3 4th Floor, Medical School, Framlington Place, University of Newcastle

    This presentation will be given by Professor Ali McGuire of City University, London.

    "Efficiency in the hospital sector: demand uncertainty and endogenous output"

    Production and cost duality will be broken if the standard assumption of independence between cost and demand does not hold. Moreover, if output is no longer deemed to be exogenously determined, but rather an partly endogenous choice variable the specification of the cost function needs to take account of this. In the hospital sector, postulating that some demand (e.g. emergency demand) fluctuates stochastically and that some output (e.g. elective demand) is endogenously determined, alterations to the standard notion of efficiency are required. This paper presents a broad theoretical framework which incorporates these alterations and discusses the implications for empirical research. Some results may, time permitting, be presented!

    Following the presentation the study group will entertain Professor McGuire to dinner. Those attending the presentation are welcome to join us for dinner at their own expense. The cost of the meal will be approximately stlg20 per person.

    In order to have some idea of numbers likely to attend the seminar, could those intending to come please let me know of their intention to do so. If you intend to come for the meal, also please ensure that you let me know by 1st October so that the necessary arrangements can be made.

    Anyone interested should contact the Organiser, Bruce Hollingsworth

    Email: bruce.hollingsworth@newcastle.ac.uk

    Tel: 0191 222 7027

    Fax: 0191 222 6746


  • Report of the Meeting in October 1997
    Professor McGuire began by suggesting that production and cost duality will be broken if the standard assumption of independence between cost and demand does not hold, depending on the holding of capacity for stochastic demand. Moreover, if output is no longer deemed to be exogenously determined (as is the norm in health care analysis), but rather an partly endogenous choice variable, the specification of the cost function needs to take account of this. In the hospital sector, postulating that some demand (e.g. emergency demand) fluctuates stochastically and that some output (e.g. elective demand) is endogenously determined, requires alterations to the standard notion of efficiency.
    It was argued that, assuming length of stay to be the same for emergency and elective cases, if uncertainty is taken into account when production decisions are made, if a hospital is operating off the production possibility frontier this may be a response to uncertainty as to numbers of emergency admissions rather than a consequence of inefficient operation. When this demand uncertainty is accounted for, the marginal costs for emergency and elective admissions is significantly lower. This may have a number of policy implications, as this uncertainty may have an impact when production decisions are made.


  • Meeting In July 1997:

    John Rousseau of the University of Texas. It will be held on July 3rd 1997, University of Bath.

    The title of Dr Rousseau's talk is:

    Performance Evaluation of Organisational Forms and Distribution Systems in the U.S. Property and Liability Insurance Industry

    A major premise of the financial theory of the firm is that more efficient structures and product distribution systems will, in a competitive market, dominate the industry. In the insurance industry, however, there are many different organisational forms (stock, mutual, Lloyds Association, reciprocals, etc.) and many different distribution systems (direct writers, exclusive agents, independent agents, mail order, etc.). This study employs the methodology of data envelopment analysis (DEA) combined with rank order statistical methods to investigate whether there are statistically significant differences in efficiency between different organisational forms such as stock vs mutual companies, and between different marketing systems such as independent agencies vs direct writing, etc. in the U.S. Property and Liability insurance industry. A sample of 1524 companies was analysed, first as eight separate subgroups -- each with its own particular organisational form and/or marketing system -- and then as eight paired groups combined in a way to allow for comparison of individual group efficiency. Results and a comparison with prior studies are presented. Findings from a preliminary investigation into the relationship between solvency and efficiency are also presented, along with their implications for state regulatory activity.

    The meeting is scheduled for 5 pm, July 3 1997 at the University of Bath. Room details to be confirmed - Please ring Zoe Evan on 01203 528220 or fax her on 01203 524 539

  • Inaugural Meeting:

    Comparative Efficiency Studies in the Regulatory Context. By Maurice Shutler, formerly Chief Industrial Advisor, Monopolies and Mergers Commission. 5 - 6.30 p.m., 7 May 1997, London School of Economics (Venue to be confirmed).

    Attendance at Study Group meetings is free and there is no need to reserve a place but it will help planning if those intending to attend did let the Study Group Chair know of their intention.


  • Report of the inaugural meeting of the PMSG Does Productivity Measurement Impact Water Company Workers In Eastern and Western Europe in a Different Way?

    Most readers will probably be as puzzled by this question as the audience was when Maurice Shutler posed, indirectly, the question at the Inaugural Meeting of the Productivity Measurement Study Group on May 7th.

    Maurice has a long experience in Productivity Measurement, most recently as Chief Industrial Adviser to the Monopolies and Mergers Commission. He was able to give his audience a good historical perspective on the current upsurge in theory and applications surrounding Data Envelopment Analysis (DEA) and Productivity Measurement more generally.

    He suggested that the origins of current methods of Performance Measurement could probably be traced back to studies in 1968 by the National Board for Prices and Incomes of the marginal costs of postal sorting offices, though these studies had been preceded by so called "scientific management" in the 30's and attempts to link wages to productivity in the 50's. The 70's saw many studies of manpower levels at hospitals, attributing unexplained variations in manpower levels between hospitals to differences in managerial efficiency. A productivity measurement method invented by Maurice Gayton of the Monopolies and Mergers Commission in 1983, employed most notably in the coal industry, was that of "expected cost" whereby a projection is made beginning ten years ago of what the costs of production would have been for a standardised product such as coal, assuming a constant resource mix and rises in the prices of inputs in accordance with their national indexes. Differences between expected and observed unit costs are decomposed into those attributable to manpower, raw materials and other causes. The studies apparently showed a large real increase in unit manpower costs in the British coal industry over the 70's, largely attributable to failed productivity deals. An even more comprehensive study of marginal costs in this industry had been carried out in the early 70's and shelved. The results were the same in 1983 but this time most of the results were published. Links between this productivity measurement study and the national coal strike in the 80's may not be completely coincidental.

    With the advent, of the Audit Commission, privatisation and regulation in the 80's, and the parallel theoretical developments of methods for productivity measurement, most notably DEA, there has been a step change in both scope and depth of Productivity Measurement. Maurice noted in particular the sophisticated studies carried out by OFWAT of the productivity of water companies in the course of the periodic review of 1994. He also noted that there are many unpublished productivity studies by OFFER.

    Interestingly, the sophisticated productivity measures carried out by OFWAT had been preceded a decade or so earlier by simpler studies by the Monopolies and Mergers Commission of the productivity of the then statutory water companies. Unfortunately no one in the UK seemed to take note of the reports at the time except for the French. The most efficient British water only companies were taken over by French interests in the run up to privatisation.

    Similar studies of the productivity of Eastern European water companies are now being carried out with a view to their privatisation. However while water workers in this country whose companies were efficient were merely taken over by French interests, productivity measurement is not likely to impact their Eastern European counterparts in the same way. Efficiency gains through downsizing manpower can only be effected in stages in Eastern Europe because of officially-set contract conditions. Therefore Eastern European water workers can be assured that French interests are unlikely to extend the same hand of 'friendship' to them as they did to their British counterparts.


Link to DEA Homepage