Quality-related costing: Findings from an industry-based research study

Engineering Management International, 4 ( 1988) 247-25? Elsevier Science Publishers B.V., Amsterdam - Printed in The Netherlands

INDUSTRY-BASED

247

CNG: FIN INGS FRO RESEARCH STU

J.J. Plunkett and B.G. Dale Department of Management Sciences, U. M. I. S. T., Manchester (Great Britain)

ABSTRACT With the support of the Science and Engineering Research Council and the cooperation of several manufacturing organisations the authors haoe carried out a major study of the collection and use of quality-related costs in manufacturing industry. The paper summarises the main limitations and omissions of the literature on quality costing and provides guidance on the most authoritative reading on the subject. Experience acquired whilst carrying out the investigations at collaborating companies is re-

iNTRODUCTlON As part of a two-year industrial research project funded by the Science s;nd Engineering Research Council (SERC), major case study work has been carried out at four companies. This has been supplemented by experience and data collected at a number of other organisations and enriched by discussions with people who have shown an interest in the research. This paper presents an overview of the main findings arising from the study.

QUALITY-COSTING LITERATURE: A SUMMARY OF FINDINGS When carrying out a literature 0 167-54 19/88/$03.50

survey, one is

0 1988 Elsevier Science Publishers B.V.

viewed and some of the major issues are discussed, including definitions of quality cost categories and elements, collection and reporting of quality cost data and the uses of quality costs. Alternatives are suggested to (or may be used to complement) the usual prevention-appraisal-failure approach to cetegorisation of quality-related costs which is so firmly entrenched in both management thinking and the literature on the subject.

usually fortunate enough to find extensive bibliographies prepared by other workers in the field. However, there is a shortage of such bibliographies on quality costing in papers and books. With the exception of Burns (1976), Jenney (1974 ) , and Campanella and Corcoran (1983)) the most fruitful sources were found to be the Production Engineering Research Association lists of quality publications ( PERA, 1983a, b), the decennial indexes of Quality Progress, and The Institute of Cost and Management Accountants reading list on the topif:. The following are a summas*y of the main point s which emerged from the search. e There is a discrete bc)dy of iiterature devoted to quality-related t5:osts. It is available from within the general quality-oriented (as

248

or industry-oriented) opposed to tradeliterature. l Given the Japanese reputation in quality management it is perplexing that only one English-language paper from a Japanese source ( Kume, 1985) came to the atten.tion of the researchers. Nor indeed were any references to Japanese papers found, though Garvin (1983 ) does quote comparative Japanese/US costs. l In most literature searches it is found that, despite consulting perhaps hundreds of references, a dozen or so encapsulate almost everything wortl: ~;:.:ir\g cn the subject. This study r was no exic~_ ,.J~‘ ?-om accountants, the contributions of Sand;&0 (1985)) Morse (1983) and Cox (1979, 1982) should be compulsory reading for any would-be collector of quality costs. To these should be ad.ded the names of Jenney (1974)) Jordan (1983), and Liebert (1968,1976) for some thought-provoking philosophy on the subject, whilst Feigenbaum (1983 ) , Blank and Solorzarno (1978)) Hallum and Casperson (1975)) Booth (1976)) Alford (1979a, b, c), British Standard 6143 (WI, 1981))and the excellent American Society for Quality Control publications (ASQC, 1970, 1977, 1980) deal best with the practical det.ails of how to do it. l The treatment of quality-related costs in the literature is dominated by prevention-appraisal-failure catagorisation of costs. Furthermore, there is a similar pre-occupation with in-house costs and little consideration of supplier- and customer-related costs. l Much of the data in the literature is of questionable value because it is not suitably qualified by detailing inclusions and methods of computation. a The measurement and collection of costs, and warnings of the limitations of costing exercises, are poorly covered in the literature. o The wide acceptance of a project approach to quality improvement largely precludes discussion in the literature of other ways of reducing quality-related costs. The use of quality costs for planning and control is wideiy noted in the iitesature but,

based on the results of this study, is rare in practice. l Notable omissions from the literature are the failure to discuss decinitions per se, despite obvious differences in interpretation which occur, and the lack of guidance on cost collection (irrespective of the topic) including treatment of overheads and valuation of scrap. In addition, costs arising from modifications, concessions and design changes are poorly dealt with in the literature. l Published definit.ions are needed in accounting terms relating to quality. l The quality-costing literature appears to be out of date because it does not reflect the changes in philosophy which have brought about a transition from inspection-based quality control to quality assurance and then to total quality management or company-wide quality control.

BRITISH STANDARD 6143 - GUIDE TO THE DETERMINATION AND USE OF QUALITY-RELATED COSTS The terms of reference for the research project included testing the applicability of BS 6143 (BSI, 1981) in manufacturing industry. This was carried out at two companies. Plunkett and Dale (1986) describe in some detail the results of this testing at one of the companies. The following are the main criticisms of the Standard. l BS 6143, “Guide to the determination and use of quality-related costs”, is a poorly abridged and edited version of its U.S. counterpart in the ASQC’s “Quality Costs - What and How” (ASQC, 1970). l The Standard does not reflect accurately the incidence and distribution of costs in the manufacturing companies collaborating in the study. The emphases in the Standard reflect the magnitude of quality activities rather than quality-related costs. The presentation of the Standard does not engender a flexible approach t.o quality costs, even though it may have been CollpCting

249

intended. o Some cost elements in BS 6143 are inappropriate to manufacturing industry and are more suited to the heavy-fabrication industry. l It is difficult to find generic terms to describe specific tasks or activities having the same broad objective in different industries or t.ypes of manufacture. This makes collation and comparisons of data from different sources very difficult. l The Standard permits reporting net costs by deduction of income from sales of scrap from quality costs. This is not good practice because it subsidises poor quality. Also, the type and quantity of scrap sold at a particular time may bear no relation to current output. o The checklists in the Standard can be useful thought-starters when collecting quality costs, but they can also act as blinkers. o The Standard should make specific recommendations on the quality-relatedness of testing and running-in operations, and on accounting of overheads and scrap.

CASE STUDY FINDINGS In this section, defini:: ions of quality cost categories and elements, collecting and reporting aspects of quality costs, and uses of cost data are discussed. Categorisation of quality costs The widespread use and deep entrenchment of the prevention-appraisal-failure (PAF) categorisation of- quality costs invites analysis of the reasons for it. This categorisation is usually done after the collection exercise for reporting purposes and adds nothing to the data’s potential for provoking action. It may facilitate comparisons with earlier data from the same source (but even this may not be valid because of the irrelation of current warranty costs, where these are included, to other current costs j. The PAF categorisation of costs is of interest and some

value to quality managers, but less SO to other functional managers on the grounds that they do not relate directly to the activities of the business. However, there are some general and specific advantages to be gained from this type of categorisation. Among the general advantages are that it may prompt a rational approach to collecting costs, and it can add orderliness and uniformity to the ensuing reports. Among the specific advantages of this particular categorisation are, firstly, its universal acceptance, secondly, its conferance of relative desirability of different kinds of expenditure, thirdly, and most importantly it provides keyword criteria to h.elp to decide whether costs are, in fact, quality related. The last-mentioned point may explain why neither Feigenbaum (1961) , the originator of the categorisation, nor the ASQC defines the term “quality costs”. Matters are judged to be quality-rela.ted if they satisfy the criteria set by their definitions of prevention, appraisal and failure. A possible alternative (cm supplementary) categorisation suggested by the study is to gather quality-related costs under headings of supplier, company, and customer. This idea came frc)m watching quality departments at work and the realisation that the PAF concept of quality activities, for all its virtue, relates more to disposition of historical quality costs than to the activities of the company. It is suggested that costs categorised under “supplier”, “company” and “customer” headings would relate closely to the way companies work. Quality costs categorised in this way would relate better to other business costs and wtuld be easier for people to identify with. However, both the business and the quality concepts of costing could be met by the simple expedient of a suppreventionplier-company-customer vs. appraisal-failure matrix, as illustrated in Table 1,.The table also indicates some major cost elements which might be included in the matrix. This proposal has the merit of new categories which closely match the business activity whilst retaining the advantages of the established categorisation.

250 TABLE 1 n.._1:+__ r*UUllb)

_^_A LU3L

-._L__-lllclll

ix

Supplier/subcontractor

In-house

Customer

Prevention

Supplier quality assurance, feedback, advice training Vendor assessment and rating Certification and accreditation Develop existing and alternative sourcing Audits and site inspection Joint quality planning

Training Statistical quality and process control Quality engineers Quality planning

Joint quality planning Field trials Evaluation of customers packing, handling and storage arrangements Market research Customer audits and inspections

Appraisai

incoming inspection Sorting Organisinp returns and replacements Inspections at suppliers site Material certification/traceability

Inspection and test Product testing Cahbration Checking procedures

Product sign-off/certification Liaise with customer inspection

Internal failure

Work costs to point of scrapping Rework costs Negotiate rework/sort prices Machining defective materials Lost production and disruption owing to defective material Negotiate reimbursements

Isolation of causes of failures Reinspection Modifications and . conc,sssions Scrap and associated costs Rework and associated costs Downgrading

Discount on goods accepted on concession Downgraded goods sold cheaply

External failure

Costs attributable to but not recoverable from supplier Complaint handling Receipt and disposal of defective goods

Analysis and correlation feedback data

of

Complaints handling Customer returns Free of charge replacements Field repairs

Warranty

Costs attributable to suppliers but not recoverable

Analysis and correlation feedback data

of

Warranty payments Warranty claim checking and negotiation

Other

Excess stocks to buffer delivery failures Preparation of specifications

Records of quality-related activities Quality performance reporting Quality costing Interdisciplinary quality task groups

Definitions of quality costs Two of the salient points which have emerged and are important to any discussion of the topic

Product liability provision /insurance

are that accounting systems do not readily yield the information needed, as it is presently defined, and that rigorous definitions of quality activit;I elements are necessary only for costing

purposes. Thus there is an apparently absurd situation of defining elements in a way which makes them difficult to cost. Elements are usually identified from specific activities or expenditures arising from non-conformance (in the broadest sense) without much consideration of the ease of costing. Given that accounting systems are unlikely to change radically to accommodate squality costing difficulties? there should be greater consideration of the accounting aspects when defining quality elements. The objective must be to re-define quality elements to align with the business activities of the company and fit in with the costing structure. Warranty cost is an example of such an element. It is clearly quality-related, it is part of the business agreement between a company and its customers, and a company must make financial provisions to meet its liabilities. Another example is the method of quality costing at one of the co-operating companies, where most of the data came directly from cost centres or labourbooking accounts. When considering definitions of quality and their susceptibility to costing, the accountant’s preference for definitions *which are constrained to meeting specification has much to commend it. Open-ended definitions such as “fitness for purpose” admits too many intangibles and makes costing more difficult. If, say, “fitness for purpose” is the quality objective, it must be met through suitable specifications and cost collectors must not be left in the difficult situation of trying to decide what parameters affect the product’s suitability for its purpose. The problem of whether testing is a quality cost arose at every company studied. Carson (1985 ) is positive that testing is about detecting defects, that it is an appraisal cost, and that there is an onus on the manufacturing department to “get it right first time”. Testing is effectively proving the fitness-for-purpose of the product in one or more respects. There may well be cases where such testing might not be necessary but is and hence incurs a quality cost. However, in many cases the state of the technology may be such that testing is unavoidable.

A manufacturer may be unable to give guarantees without testing the product. He may be unable to get insurance cover without testing. In the end, the decision whether to test or not may be taken out of his hands, whatever he thinks he can achieve without testing. In such circumstances should testing be regarded as a quality cost? And if it is, is it an appraisal or a prevention cost? An aspect of definitions which arose and which is worth discussing briefly is the sensitivity of costs to changes in quality. It was noted at one of the companies that. despite dramatic improvements in quality there was little evidence of corresponding changes in quality costs. The matter arose again in a different form at another company where they were looking for a cost reporting system which reflected quality performance. Unfortunately there are many costs which do not change pro rata with quality improvement. These costs are often large and obscure those costs which do change with quality. The answer to the problem seems to be to identify and isolate those parts of the cost which change (or will ultimately change) pro rata to the quality improvement and to report them separately, if necessary, perhaps as projected savings. Finally, a new set of definitions is needed to help determine costs associated with concessions and engineering changes. The study has shown that they are major quality activities but little is known about their costs, because they are not picked up using the usual quality cost checklist. This would help to focus attention on these activities to make people more aware of them, and to make accountable those people who are responsible for incurring the costs. Collection of cost data The need to get the purpose of a cost collection exercise clear at the outset is of considerable importance to the success of the outcome. Possible strategies range from measuring and monitoring all quality costs to costing only specific quality improvements projects. Total

252

quality costs inevitably include large immutable costs, and the powerful arguments for measuring only costs which change tend to erode the case for collecting total quality costs. An acceptable compromise is to carry out occasional total cost exercises but to monitor regularly and emphasise only those costs which are likely to change. The basis of the argument supporting this view is that it is not necessary to know all the costs to be sure, for example, that quality costs are increasing. Another argument, leading to a similar conclusion, is that if a company is !ooking to improve costs while maintaining quality (or vice versa) it needs measures of quality and cost performances, and of their relationship to each other. Further, it needs only quality maintenance costs and those costs which can or do change with quality. Whilst such arguments are plausible and superficially attractive, they imply that some key costs are sensitive to quality changes and that the relationship is known and understood. This study indicates that this is far from being the case. One of the important decisions on strategy, to be taken at the outset, concerns the assignment of accountability. Deciding in advance who should be accountable for what may produce a very different cost report to one based on the usual quality-based criteria. Also on the subject of accountability, the production fu: rtion is the most closely measured and acountable group in manufacturing organisations and, hence, it is usual for cost collecting to start there. It is important to make sure cost collection does not end there, for although the manufacturing function has a prime responsibility for quality, there are many others who contribute to quality non-corformance, failure, and quality costs. When actually collecting quality costs, it is sometimes easy to lose sight of the fact that the task is primarily a cost collection exercise, and that cost collection exercises have other, different criteria which are sensibly independent of the cost topic. Plunkett and Dale (1985) suggest as suitable criteria: purpose, relevance, ease of collection, size, accuracy, completeness, po-

tential for change, recording and presentation, and uses. A set of back-up criteria like these, which are independent of the cost topic, can often provide useful ways out of the dilemmas about whether particular activities and costs should be included in a costing exercise. Another obvious but often overlooked fact is that costing systems only translate other information into costs (i.e. the subject matter needs to be recorded in some form before it can be picked up by a costing system). It follows that the potential for costing at a company may be gauged from the sophistication of its management information systems, and also that overambitious pursuit of detailed cost information will of necessity proliferate recording and analysis of underlying information. Over-ambition or over-zealousness may also prompt quality managers to try to maximise costs, not necessarily for self-aggrandisement, but to try to create a financial impact. Unfortunately, it can backfire and they may find themselves stuck with costs which are independent of qualAcy considerations, and over which they have no control. It is not always easy to disown costs after one has claimed them, especially if ownership is in a “grey area” and noone else wants them. At the other end of the spectrum, care must be taken not to concentrate only on what is already known with a view to refining it. However, if refining costs is the object of the exercise, it is probably much more advantageous to refine known large costs than to identify and quantify sources of small costs. An essential precursor to collecting qualityrelated costs is knowledge of the company and its processes. Whilst this is clearly good advice to academic researchers, it should not be imagined that it is an unnecessary consideration for companies carrying out their own quality cost exercises. Costing requires technical and accounting knowledge, and few people at the right level in the organisation are likely to have all the knowledge that is necessary. If accounts people try to do it alone, they are likely to miss a lot, or even be misled by people with axes to

253

grind. If technical people go it aibne, their costs may lack credibility because they may, for example, fall to discover costs which tidy-minded accountants have tucked away out of sight. Thus, costing should be a joint exercise. The findings from the study indicate that the initiator and driving force behind quality costing programmes is usually the quality manager. However, it is our view that only costs produced or end.orsed by accounts departments should be used in reports. Such costs will have greater acceptability and are more likely to be compatible and consistent with other cost efficiency measures. It should be remembered too that management accountants are always under pressure to produce all sorts of costs and, if quality costing is to get off the ground in a company, they need to be convinced that there is some worth in the exercise. Initiatives from accountants seem to be prompted only by high costs of inspection or scrap appearing in labour or materials cost analyses, respectively. IBecause cost data is expressed in figures which are sometimes very precise, there is a danger that cost reports may give an impression of accuracy which is not warranted. The true accuracy of costs depends on the underlying data. Knowledge of how these data are obtained and the purposes for which they are used may give good indications of the accuracy which may reasonably be sought or expected. Resorting to underlying data can also have the advantage that corroborative information may be available, though corresponding costs may not be. The accuracy sought has a strong and direct influence on the amount of work involved in the task and on the credibility of the outcome. It is not usually practical LOdecide in advance of the task the level of accuracy at which to work or which is attainable. Rather, the accuracy possible is usually determined by factors outside the control of the cost investigator. Against the background of difficulties with definitions of knowing what should be included amongst quality-related costs and the problems of gathering some of the costs, there seems little point

‘n pursuing accuracy even where it is known to be obtainable. On the other hand, costs must be accurate enough to be credible. Nevertheless many companies are content with quality costings which are 2 10% and, seemingly, the greater the cost the greater the tolerance. On the matter of accuracy, guesses are at best useless. The danger of course is that the guesses are either ascribed the same credence as other figures in the report, or they may be identified as a weakness through which the report may be discredited. They will certainly not be kept in their proper perspective. As a final point it is important to remember that in addition to having the necessary mechanisms in place for collecting quality-related costs, it is also necessary for senior managers to have the will to carry out the collection process. Reporting quality-related

costs

Repsrting of quality costs was found in the main to be a sub-section of the general reporting of quality department activities and as such loses its impact. For maximum impact, quality costs should be identified in an organisation’s cost reporting system. Overall, the standard of reporting of quality costs in the companies studied was poor. Good standards are essential if reports are to make an impact and provoke action. Senior managers are like everyone else in wanting easy decisions to make. Having costs tangled up with t+echnical information makes the data less clear than it could be and often provides an excuse to defer a decision. The problem for the senior managers should not be to disentangle and analyse data in order to decide what to do, it should be t5 decide whether to act, choose which course of action to pursue and ensure provision ol; necessary resources. Problems, possible solutions and their resource requirements should be presented in the context of accountability centres which have the necessary authority to execute the dCcisions of the senior management. The collection and presentation of costs should provide ai; E;he outset for all those ele-

254

ments and cost sources that are worth collecting. The creation of a quality-related cost file integrated with existing costing systems but perhaps with some additional expenses codes should not present many problems. Collecting the data will probably be much more difficult. Those which come from within the quality department may be easy to obtain but those from other departments may be difficult. Nevertheless provision should be made in the file for COGletting such data, even though it may take a long time to obtain satisfactory returns on a routine basis. A popular view among quality practitioners is that quality cost reports should indicate the origin of failure costs by department (e.g. production, engineering, marketing, etc.) in the hope that adverse publicity will provoke remedial action. Unfortunately, it may also antagonize departmental managers so that they become uncooperative in providing information for the report. It may even result in deliberate obscuring of evidence of quality performance. Emphasis on improvement opportunities rather than attribution of failures may help to avoid such difficulties. An important consideration in the presentation of quality-related costs are the needs of the recipients and it may be worth presenting information in several different formats. For example, weekly reports of costs and scrap and rework are of greatest value to shopfloor supervisors; monthly reports of total costs, highlight.ing current problems and progress with quality improvement projects, would be most suitable for middle management; and total costs and costs to act on are needed by top management. While selective reporting of this type has its merits, it should be done against a background of the total cost of qu?lity. It is important to collect and display all those costs which are available, and also to indicate the existence of relevant costs which can not yet be quantified. It is important, because reporting only part or’ the costs can be very misleading, and reporting the existence of unquantified costs keeps them in view and encourages attempts to measure them.

The time-lag between manufacture and warranty payment is often so long that warranty payment should not be reported in the same context as current quality cost. If the effects of the time lag and ‘.he effects of price changes, inflation, exchange rates, etc. are added, there can be no doubt warranty costs ought to be reported as a separate category of quality cost. In the context of other (non quality) costs it should be reported as a loss and related to other losses and profits. Use of quality-related

costdata

The level of sophistication of quality costing differed at each of the four companies where major case-study work was conducted. However. they were all looking for better ways of using quality cost data. Very briefly: l Company 1 just wanted to know how they stood in relation to other companies. l Company 2 was looking for a uniform basis for comparing six divisions of the company. l Company 3 was trying to highlight performance, and contribution to the quality record, of people at all organisational levels. l Company 4 was looking to costs as a way of motivating and achieving improvement. It is interesting to note that these are not so much uses of cost information but situations that they would like to achieve through the use of cost information. By and large questions about uses do not elicit answers which are specifically cost-oriented (e.g. set cost reduction targets, establish cost efficiency measures). Three of the companies did not use qualityrelated costs because they did not have much quality cost information routinely available to them. Whilst this would obviously handicap them in setting overall cost reduction targets, it Jid not hinder them from pursuing quality improvement projects which had been identified from technical or quality control information, and from using approximate costs to justify and monitor the projects. In some industries (e.g. large, heavy fabrications) 9 it is difficult to build up a history of quality costs for progress to be measured. Companies in these kinds of

industries may onl;l use costs in the context of improvement projects. Taken overall, quality improvement projects appear to offer the quickest route to useful exploitation of cost data. Utilisation of total quality costs takes much longer. The study revealed that ranking of competing projects by different cost bases and by numbers gave very different indicators. The preferred base depends on factors and circumstances that may change with time or with a company’s business situation. Different indicators may require very different amounts and types of resource. Perhaps the most valuable use to be made of cost data is to help companies to decide how, when, and where to invest in prevention activities or equipment. But even if full cost data were available it may not be useful for this purpose, because the fundamental economics of prevention (cr appraisal) do not seem to be understood. Indeed, whenever investment in prevention expenditure was discussed, an implicit expectati.on arose of a large return on the investment. While recognising that companies are in lusiness to make profit and that there may be Tompetition for limited resources, the payback -upected (and sometimes proffered) from investment in quality seems unusually high. Since higln returns are usually associated with high risk, one wonders whether as a general rule the benefits from investment in quality are seen in business circles as being problematical.

CONCLUDING

REMARKS

A measure of the status of quality costing in the companies studied is that it’ is not featured in their quality manuals. The research has confirmed that quality costs can be a useful management tool but there are many complexities and difficulties to be overcome before quality can enjoy the same status as other major business parameters. Deciding which activities should be included under the quality-related cost umbrella is by no

means straightforward and there are many grey are723, $uality managers tend to include under the quality umbrella costs that are difficult to justify as quality-related and over which they have no control or influence. Another area of difficulty is deciding whether some activities, usually Of a setting-up, testing or running-in type, are ql-ality activities or an integral and essential p .t of the manufacturing operation. These costs often can be substantial and can alter quite markedly the relative proportions of quality-related cost categories. There are also factors which serve to ensure the basic utility of the product, guard against errors, and protect and preserve product quality. Examples are the use of design codes, preparation of engineering and administrative systems and procedures, capital premiums on machinery, document and drawing controls, and handling and storage practices. Whether such factors give rise to costs which may be regarded as being quality-related is a matter for judgement in individual cases. Quality activities of manufacturing companies, and their cost reports based on local perceptions of quality-related costs, are much more supplier and customer oriented than the usual prevention-appraisal-failure checklists suggest is the case. Categorisation of quality-related costs under supplier/in-house/customerrelated costs may produce elements and reports which are more closely aligned to the way the business operates than does the usual prevention-appraisal-failure categorisation. Whichever method of categorisation is adopted, there is little or no point in routinely gathering and presenting costs that do not change. Hence, before setting up a cost collection system, it is advisable to examine the potential for change of a cost element in both absolute and relative terms. The inclusion of fixed or immui,&ie costs also has the effect of reducing the sensitivity of costs to performance changes. The size of quality-related costs, both absolute and relative can become grossly distorted under some accounting practices. One such practice is the inclusion of a proportion of over-

256 TABLE

2

Do’s and dont’s for a quality

cost collection

exercise

Get the purpose and the strategy clear at the start. Report costs produced or endorsed only by accounts departments. Get data and costs from standard data wherever possible. Seek independent corroboration of any data which is doubtful. Start with failure costs. Consider appraisal costs as a target for cost reduction. _ Consider ease of collection and start with the cJa+st. Refine large costs rather than quantify small t nkn 3wn costs. Concentrate on costs that do or can change wi; h quality. Analyse and report costs clearly in a busine: = context. DON’T (or engineering) help. Go it alone - seek accounting 0 Be too ambitious - start small. 0 Expect too much from the first attempt. e Agonise over relatively trifling sums. e Use guesses - not even “informed’ guesses. (b Make comparisons unless you can guarantee comparability. 0 Assume straightforward opera!ions will be easy to cost. l Forget that prevention is the most difficult of the categories e Deduct income from scrap from quality costs. 0 Include overheads in labour costs. e 0 Concentrate exclusively on what is already known.

head costs in some labour costs. Special care also needs to be taken when obtaining data from differences in provisions; accounting practices in making provisions for liabilities may obscure actual expenditures in a given period. An interesting observation made during the course of the research is that the circumspection of accountants about the ease of collection of quality-related costs contrasts sharply with the blithe optimism of quality professionals. All the companies studied were keen to know if their quality performance in terms of qualityrelated costs is good, bad or indifferent, but comparative data is scarce. However, it is dangerous to compare quality cost data between different companies and industries unless their accounting practices and cost computations are known. In general quality-costing exercises in manufacturing industry produce activity-oriented elements that are appropriate to the technology, the business, and the organisation.

.-------_L

to cost.

As a final concluding point, Table 2 states briefly a few do’s and don’ts which may help organisations to avoid some of the difficulties and traps typically encountered in a quality cost collection exercise.

The authors acknowledge the financial support of the Science and Engineering Research Council through their research contract GR/C33475. They also wish to thank Mr. Jim Morrison for drawing to their attention the paper by Kume (1985).

REFERENCES Alford, R.E., 1979a. Quality costs I. Quality, (August): 14-K

Where to start.

Part

257

Alford, R.E., 1979b. Quality costs - Where to start. Part II. Quality, (September): 70-71. Alford, R.E., 1979c. Quality costs - Where to start. Part III. Quality, (October): 40-42. ASQC, 1970. Quality costs - What and how American Society for Quality Control, Milwaukee, WI ASQC, 1977. Guide for reducing vendor costs . American Society for Quality Control, Milwaukee, WI ASQC, 1980. Guide for managing vendor costs American Society for Quality Control, Milwaukee, WI. Blank, L. and Solorzarno, J., 1978. Using quality cost analysis for management improvement. Ind. Eng., lO( 2): 46-51. Booth, W.E., 1976. Financial reporting of quality performance. Qual. Prog., (February) : 14-15. BSI, 1981. British Standard 6143. Guide to the determination and use of quality related costs. British Standards Institution, London. Burns, C.R., 1976. Quality costing used as a tool for cost reduction in the machine tool industry. Qual. AssuranLc, 2 (1) : 25-32. Campanella, J. anu ’ Corcoran, F.J., 1983. Principles of quality costs. Qual. Prog., XVI (4) : 17-22. Carson, J.K., 1985. Quality costing - A practical approach. Int. J. Qual. Reliability Manage., 3 (1): 54-63. Cox, B., 1979. Inierface of quality costing and terotechnology. Accountant, (June) : 800-801. Cox, B., 1982. The role of the management accountant in quality. Qual. Assurance, 8 (3 ) : 82-84. Feigenbaum, A.V., 1961. Total quality control. McGraw Hill, New York, NY, 1st edn. Feigenbaum, A.V., 1983. Total Quality Control. McGraw Hill, New York, NY, 3rd edn. Garvin, D.A., 1983. Quality on the !ine, Harv. Bus. Rev..

( September/October) : 65-75. Hallum, S.A. and Casperson, R., 1975. Economy and quality control. In: Proceedings EOQC Conference, Venice, Italy, pp. 11-22. Jenney, B.W., 1974. Motivation for quality. Qual. Eng p 38(l): 5-7. Jordan, P.E., 1983. The relation between quality analysis and quality strategy in engineering companies. Qual. Assurance, 9 ( 3) : 59-62. Kume, H., 1985. Business management and quality costs: the Japanese view. Qual. Prog., (May): 13-18. Liebert, F.P., 1968. Guidelines on the gathering and implementation of quality costs. Qual. Eng., 32 (2) : 39-43. Liebert, F.P., 1976. A standard approach to quality costing. Qual. Assurance, 2 (2 ) : 53-54. Morse, W.J., 1983. Measuring quality costs. Cost Meas., (July/August) : 16-20. PERA, 1983a. Quality assurance: a select bibliography. Production Engineering Research Association, Melton Mowbray, Leicestershire, U.K., August. PERA, 1983b. Quality assurance: a select bibliography Supplement. Production Engineering Research AssoMelton Mowbray, Leicestershire, U.K., ciation, December. Plunkett, J.J. and Dale, B.G., 1985. Some practicalities and pitfalls of quality-related cost collection. Proc. Inst. Mech. Eng., 199B (131) : 28-33. Plunkett. J.J. and Dale, B.G., 1986. An artless attempt to collect quality-related costs. In: Proceedings of the first National Conference on Production Research, Kogan Page, Londan, pp. 416-425. Sandretto, M.J., 1985. What kind of a cost system do you need? Harv. Bus. Rev., (January/February) : 110-l 18.