MCD Contents


The following paper was submitted to the Select Committee on Public Administration in December 2002 and was placed by the Committee in the House of Commons and House of Lords Libraries as Paper PST19.




Comments to Public Administration Select Committee


Harold Beck


There appears to be much misunderstanding about targets.  A target is only one part of a managerial control system.  It is essential to consider all parts of that system.

The following notes were compiled over a decade or more of participative sessions on managerial control conducted with practicing managers from industry, commerce, voluntary bodies and the public services.  The public services included local authorities and health authorities in which the author was at the same time active as an elected or appointed member.

A managerial control system comprises four elements:-


All four elements are important.

Managerial control is:-

MODIFYING ACTION to reduce (to zero if possible) the difference between the PERFORMANCE AS MEASURED and the PERFORMANCE REFERENCE.



An experiment demonstrating the four elements - balancing a stick on a hand.


Vertical - sensed from doorframes, right-angle to horizon etc.


Eyes measures stick position. Close eyes and measurement ceases - comparison cannot be made and control action cannot be taken.


Brain computes angle of stick to vertical.

Control Action

Hand moves to reduce angle to zero

In a closed loop system there must be consistency between the elements.


There is much in the literature about REFERENCES (objectives, targets etc) but little about MEASUREMENT, COMPARISON and CONTROL ACTION.  Looking at the elements in more detail:-

REFERENCE   (Criteria      Objectives     Plans     Budgets     Policies    Methods   TARGETS)







Financial Ratios - externally derived eg how other companies are doing





Prestige     Impartiality     Job Satisfaction


Recruitment procedures



Expenditure    Staffing


Forecasting      Appointment Interview     Committee Examining Proposal


Atmosphere    Nuances


Continuous measurement too expensive and/or disturbing so measure only now and again


Consider financial, social & other costs of measurement


Deviation Information

Variance Statement

Management by


Setting limits to deviations to focus attention on problem areas


Action Choices

Improve Performance (as measured)    Revise Budget/Change Deadline or both

Open-loop system

Economy drive

Cost Effectiveness

Loss of tools from unlocked workshop less costly than repairs to locked workshop.


Dummy cameras deter shopstealing


Effect on control system effectiveness



Expedient Simplification   This is a common error i.e. control only what is easy to control, forget the rest.  Example: choose only quantitative targets and measurements.

Science compared with Human Activities  Lord Kelvin of Largs (William Thomson) said “to measure is to know", "if you can not measure it, you can not improve it." and “when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind.”  His remarks were made in the context of physical science.  Measurement of human activities is far more complex yet we often proceed as though the same scientific approach is completely valid.

Again, arising out of Heisenberg’s work and quantum mechanics it is said “all measurement disturbs”.  The disturbing effects of measurement on human activities are of a different kind and of far greater significance - and none more so than when what is measured and how it is measured is known to the people or organisation being measured.  Example:  School Inspection - "why don’t you tell us in advance what you are looking for?"

Be creative or ingenious in measurement  Example: a factory manager was concerned that though machine operatives were switching off their machines and departing promptly for their tea break they were drifting back well after their break had ended.  Instead of employing time and motion study people to measure each operative he had an electricity power consumption recorder installed in his office to show the different patterns at the beginning and end of the tea break, which he drew to the attention of the shop foreman.

Keep under control or prevent getting out of control rather than control absolutely  Thus for example the factory manager in the above example got the foreman to admonish the operatives from time to time, when he saw the return pattern becoming unduly extended.  He did not seek a return pattern as ‘sharp’ as the departure pattern.

Temporary Control  Set targets, measure, compare, take control action, publish results then abandon that particular control system and move to another area or type of control.  If necessary revive the first control system from time to time.

Every way of seeing is a way of not seeing  When setting up one system to control a particular parameter it is worth remembering that another parameter may be obscured or affected.

Precision not supported by Accuracy  The average of several integer scores of 1 to 5 are very often calculated to 1 or even 2 decimal places and the resulting score is then used to confirm or deny a point at issue.  The precision or sensitivity of the resulting score is not warranted by the inaccuracy of the original integer scores. 

Caveats and qualifying assumptions are forgotten  The assumptions and reservations associated with each element of the control system must be linked to evaluation of results.

Accountability  An individual or group (such as a Department) cannot be held accountable for reaching targets if he/she/they cannot take control action.  It is also favourable to accountability if she/he/they accept the target and methods of measurement and comparison.

HVB 9 Dec 02



MCD Contents