Introduction: Setting the empirical and conceptual scene
Why downsize?
Why a golden handshake?
Possible new approaches and their limitations: How much to downsize?: Determining the magnitude of the downsizing
How to downsize?: To set preconditions or not to set
What to do with the wages saved from downsizing?
The Kenya example
The golden handshake: A grant or a loan to departing civil servants?
To give incentives or to dismiss
Other implementation issues: Alternatives on how to set up the payment system for the golden handshake
How to redeploy public servants to the private sector?
Conclusions
References
Public Admin. And Dev., Vol.18, 1998.
JOHN MACGREGOR
STEPHEN PETERSON
CLAUDIO SCHUFTAN
schuftan@gmail.com
Abstract
In recent years, more and more calls are being heard in a growing number of developing countries to downsize their civil services. It is argued that downsizing is needed, because of the increasing shortfalls in government recurrent and development budgets. This situation results in under-utilised, under-funded staff and often in the siphoning-off of donor funds in the development budget for recurrent expenditure.
The main problems addressed in this article are why should and how can the civil service in developing countries be downsized. The questions of how much to trim the bureaucracy and how to redeploy redundant public servants in the private sector are addressed as well.
The article examines alternative strategies for significantly downsizing the civil service. It is contended that problems in this area are indeed common to many developing countries.
Various golden handshake options for civil service leavers receive particular attention. It is suggested that ECONOMIC JUMPSTART is a better term than golden handshake to characterise the incentives package offered to induce staff to accept voluntary redundancy reductions.
Civil service reform is not presented here as a cure-all remedy for all developing countries’ ills. The article primarily makes a number of mostly untried but never-the-less attractive suggestions that bring in some fresh thinking into this difficult issue of downsizing the bureaucracy in developing countries. Paths and avenues worth exploring when starting to design civil service trimming operations are presented, including some of their limitations.
The point is finally made that this type of downsizing is overdue in many places. The article should be considered as a contribution to demystifying the process of downsizing the civil service in developing countries.
Introduction: Setting the empirical and conceptual scene:
In recent years, a growing number of developing and developed countries have been attempting to downsize their civil services. (de Merode, 1991; Langseth, 1995; Synnerstrom, 1995; Balasz, 1996; Zussman, 1996; Panov 1996; World Bank, 1997) It has been argued that this trimming is needed, because of the chronic and worsening government recurrent and development budget shortfalls. (Fallon and Pereira da Silva, 1991) These shortfalls are most acutely seen in operations and maintenance (O + M) expenditures and often lead to rising government borrowing to cover the public deficit. This core fact makes oversized bureaucracies increasingly unsustainable.
As the civil service staff complement has risen to absorb rapid growth in the labour force, the O + M portion of budgets has grown much more slowly, not even keeping up with inflation. This situation invariably results in idle staff with no funds to carry out their functions and often in the siphoning-off of donor funds from the development budget to cover O + M expenditures. For some time now, these contentions have been true for the majority of developing countries. (Ozgediz, 1983; Nunberg, 1989; Toole, 1989)
Donor policies often condone oversized bureaucracies. For example, the World Bank’s quick disbursement balance of payment support policies do allow governments to borrow while still sustaining untenable large civil services. (Nunberg and Nellis, 1990)
The main problems being addressed here are thus why should and how can the civil service in developing countries be downsized. In addition, the questions of how much to trim the bureaucracy and how to insert dismissed public servants in the private sector get due attention. Moreover, it is recognised that downsizing also has to deal with redefining the functions of a trimmed bureaucracy, and with providing the right incentives for the downsizing exercise to be acceptable.
These and other issues are dealt with here using some data from Kenya -where the three authors lived and worked in the early 1990s. Kenya itself (with a population of 24 million and a civil service of about a quarter of a million in 1993) had not embarked on a civil service reduction plan at the time, though. Therefore, the article examines alternative strategies that would have been available. The problems of Kenya’s civil service -i.e. excessive staffing at lower grades (grades A-G in Kenya), wage compression, low output and low morale- are common to many developing countries. (Adamolekum, 1989)
This is not a case study of how the bureaucracy was trimmed in Kenya; examples from Kenya are utilised in the article only to illustrate what is proposed and to highlight general principles applicable to downsizing. The options explored are thought to be relevant to other countries as well. Suggestions are also made about possible new approaches to downsizing -including donor-financing alternatives. No attempt is made to make recommendations or to prioritise actions in the process. This because, when trying to trim the bureaucracy, local decision-makers often need assistance to develop good arguments for downsizing -mostly with only a minimum of hard background evidence available to fully legitimise any particular approach being proposed. Therefore, the arguments presented mostly set new boundaries for choosing more specific courses of action in different contexts.
Why downsize?:
There are several sound arguments for civil service downsizing, the main being: to reduce fiscal deficits and thus free up domestic resources for the private sector, to reduce the effect of the superfluous staff on management time and overhead functions (Fry et al, 1988), to improve public sector productivity by tying personnel levels to adequate and sustainable O + M support (Quah, 1986; Prokopenko, 1989; Binder,1990; Ban et al, 1992), and to limit the role of the state to those tasks that cannot be adequately, willingly or profitably performed by the private sector.
As a result of the above, it is hoped that downsizing will better balance the private and the public sectors and foster an economic environment more conducive to the achievement of development goals.
There is another point of view to be pondered, though, when looking at large civil service complements. One could argue that civil servants are really more ineffective than unproductive. Given that their salaries are so low, the fact that their limited inputs result in meagre outputs does not exactly fit the definition of low productivity. From this perspective, civil servants may actually and paradoxically be considered reasonably productive in the strict sense of the word. They may be doing the wrong mix of things, though, rendering them ineffective. What this means is that for a government to employ low skilled civil servants may be a comparatively cheap social welfare function. But such a form of social subsidy has some undesirable effects: for example, large complements slow down bureaucratic processes and consume utilities, commodities and other government resources.
What remains to be proven in downsizing then is whether dismissed civil servants’ productivity will have a greater positive effect on the national economy once they are employed outside government at market values of labour.
Why a golden handshake?:
Strictly speaking, we are not here talking about a golden handshake -a term coined for the ‘Fortune 500’ managers; the civil service will never be able to offer anything close to a ‘golden handshake’. We are talking more about an ECONOMIC JUMPSTART or an incentive for redundant staff to voluntarily leave the civil service and start productive lives that better contribute to the growth of the economy. (Landau, 1969) The economic jumpstart benefits both the individual and the economy. It is in this sense that our use of the term golden handshake should be understood.
In addition to financial incentives to leave the service, the procedures used for downsizing do not exclude the eventual mandatory dismissal of some staff (see below). Moreover, it is difficult to predict whether such an economic jumpstart will also affect the performance of the civil servants left behind, i. e., whether downsizing can make the government machine work better after trimming.
Large-scale dismissal of civil servants is socially and politically unacceptable and is simply not feasible. Attrition can probably downsize the civil service only about 4-5% a year; this rate may be inadequate. (As our calculations for Kenya show below, double those numbers are needed roughly to bring the size of complement to desirable levels in about three years).
Variants of golden handshakes have been tried in several developing countries, but mostly with limited success. (de Merode, 1991) It is contended that golden handshake exercises conducted in those countries have still left many innovative options untried; we will explore them here.
When considering golden handshake options, one has to remember that it is not only the money the civil servants will ultimately consider, but also security; this may, ultimately, be the more important factor.
It seems that any downsizing scheme will work better if the incentives package offered is at least as attractive (or slightly worse, but with a perceived future potential) as staying on for a long civil service career. Although designing and costing such a package may not be easy, a few methods may be suggested.
The benefits offered to civil servants in a downsizing exercise are a national issue. Therefore, one has to develop solutions that are compatible with what exists in the country. Benefits can conceivably be better than those found in the rest of the labour market, but not exceedingly so. One can definitely not give generous benefits to redundant civil servants being asked to leave when others -in a national context- get nothing; it is a matter of finding a balance. (Synnerstrom, 1996)
Possible new approaches and their limitations: How much to downsize?: Determining the magnitude of the downsizing
The decision to downsize is a political one. Once the decision has been taken, two issues emerge: how much downsizing is needed?, and how fast does it have to be implemented to have the desired measurable impact?
In addressing these issues, one must keep in mind that each country situation will be different, and also that the final target may matter less than establishing the speed of the desired change. There are simply no generally accepted scientific methods to calculate the number of positions that should be cut; downsizing is more an empirical, pragmatic and political process.
Therefore, what measures are there for deciding how far to downsize? First, one has to make sure that the government is not borrowing from its Central Bank to meet the payroll. Second, one has to define the ideal mix of salaries (emoluments), O + M and capital investments (CI) desirable for a new government budget. Third, one has to choose the best plan of action that will bring one closer to such an ideal mix.
Looking at existing ratios between salaries, O + M and CI is, therefore, the point to start to define the above ideal mix for the specific local situation. Given understandable differences, a ministry by ministry comparison of such mixes makes more sense. Conversely -in the case of the health sector for instance- it is possible to compare the same ratios between the non-governmental (NGO) sector with those of the government’s health sector in the country. This can help come up with a more rational mix of salaries, O + M and CI.
The value of a methodology that proposes a better mix of percentage personnel costs, O + M costs and CI is that it helps determine a first downsizing target. Several avenues seem possible to do this. One could:
Look for the best practice in other countries that have trimmed their bureaucracy and then
– adapt it to the local conditions.
List the percentages of personnel/O + M/CI in government budgets for a number of countries; thereafter
– rank them from highest to lowest for percentage spent on personnel;
– determine the median and quartiles in that ranking, placing one’s country’s percentage in perspective with others, and finally
– choose a goal for downsizing using a given other country in the list as a reference.
Use a Delphi technique (that narrows down dissenting views of a group of experts) with a group of ministry personnel that has worked there for 10-15 years; thereafter
– agree upon a base year for which there is consensus that that was a good year in terms of a more balanced fiscal budget;
– look at the percentage mix of P/O + M/CI -or at least the O + M expenditure per civil servant- in that year. Looking at these ratios has to be done by sector (by ministry), because it is bound to be different in each, especially for the percentage spent on personnel;
– apply that percentage of personnel costs to the current budget as a goal to set the desirable cut-off point for downsizing. Alternatively,
– estimate the number of civil servants that would be in service if their number would have grown at the same rate than the population (or the overall labour force) since that last year with a good, more balanced fiscal budget.
Use a zero budgeting approach based on redefined new functions of the bureaucracy and then
– look at what these newly defined tasks are and how many people would be needed to perform all of them, for example in a ministry. (This is the more accurate, but difficult avenue to follow to reach a quantitative goal for downsizing. It is a cumbersome task and has the drawback that it assumes all individuals are equally productive).
Either of these methods could be used to determine in a rough way a more appropriate size of the bureaucracy and a quantifiable goal for downsizing. An example can illustrate things better. A study of the ratios of labour to O + M in two ministries in Kenya was carried out by one of the authors (S.P.) and when looking at the Recurrent and Development budgets of both, it was recognised that, in their work, the staff was using O + M resources regardless whether the same came from the Recurrent or the Development budget. It was then calculated that the preferred labour to O + M ratio for the Ministry of Agriculture would be 10:7 and 10:14 for the Ministry of Livestock Development. To bring that expenditure composition between complement (personnel), O + M and CI in line with these ratios, personnel emoluments should have been cut by 31% and by 55% respectively in the budgets of both ministries at that time. This type of calculation thus proved to be useful to set a rough goal for downsizing.
A caveat seems to be in order at this point: A more ideal mix of personnel and O + M is, per-se, no guarantee of a more efficient administration! The improvement of overall managerial skills and particularly those of talented managers will still be required. In other words, by itself, downsizing does not improve service delivery. It merely produces a civil service that is smaller, but just as efficient or inefficient as before.
In fact, a number of efficiency-oriented measures need to be taken, because downsizing per-se is not a good enough solution to the ills of the bureaucracy. (Wallis, 1989) Zero budgeting techniques may be useful in this whole exercise, and it is likely that external technical assistance may be needed to apply them.
Therefore, downsizing also has to do with the function of the state. Before and during downsizing, these functions need to be reviewed, leaving government only those responsibilities that private sector cannot do or is uninclined to do. The role of the government has to be redefined to fulfil mostly essential services leaving to other sectors what they can do better. Redefinition of functions has to come before a new look at the composition of government expenditure is attempted.
How to downsize?: To set preconditions or not to set
It is safe to assume that most developing countries’ governments cannot afford the total costs of golden handshake schemes to trim their bureaucracies. Donors can and should play a role in inducing selected governments to move towards downsizing their staff by providing the needed assistance to finance such schemes. Nevertheless, it is likely that they will want some assurances prior to stepping in with such assistance.
Among the preconditions they are likely to insist on are the following: military budget trimming (including defence capital investments) and use of the military for social service functions; when applicable, decreasing the number of ministries and privatisation of selected government activities and/or services.
Two additional preconditions would seem rather predictable: no growth in the civil service once the cutbacks have been effected (partly manifested as a hiring freeze) and savings from the downsizing not being used to write-off Treasury accumulated deficits or debt.
The simultaneous hiring freeze to be enforced while downsizing (no new hiring, no re-hiring of lower grade cadres) is to be understood, not in absolute terms, but rather in terms of ceilings in the budget for total personnel costs, broken down for each job category and group. The freeze will have to keep an ethnic (tribal where applicable) and gender balance in the little hiring that would be done, perhaps correcting for imbalances that the downsizing may create. Existing arrangements for recruitment controls need to be examined and tightened if necessary. When applicable, the posts of the civil servants who take advantage of the golden handshake will be deleted from the establishment, unless individual posts are considered indispensable.
In addition to downsizing, governments could aim at establishing a clean wage for civil servants which replaces the present system of base salary plus assorted allowances. This id due to the fact that it is the total emoluments bill that needs to be considered to work out civil service compensations. One crucial move in this direction is for developing countries’ governments to drop housing allowances and divest themselves from residential real estate, at least in urban areas. This move should probably not be a donor precondition as such, but its implementation should be carried out at the same time downsizing is taking place.
Before trimming, a census of workers will be needed in each government institution to have an accurate head count of all active staff and to get rid of ‘ghosts’ in the payroll. Existing payroll fraud controls will be examined and tightened whenever necessary. The authors know of such exercises having been carried out successfully both in Uganda and Kenya.
The experience from Kenya shows that clear accountability and oversight of staff complements is an obvious precondition, not only for donors, but for downsizing to work at all. Proof of periodic checks being carried out on electromagnetic records of the government payroll to catch duplications, inflated salaries, duplications of bank account numbers and other irregularities, must be carried out initially and need to be repeated at least yearly thereafter. The above has to be linked to the clean wage recommendation above; if not, cheating can occur in the allowances part of the paycheque.
Finally, no matter what downsizing and hiring freeze scheme is followed, it will have to go hand in hand with powers to effect the redeployment of staff in the country.
What to do with the wages saved from downsizing?
Having clear guidelines on what to do with the savings realised from downsizing is a critical issue. The most attractive option seems to be to put these funds right back into the budget (not allowing the Treasury to trim or balance the nation’s budget at the expense of downsizing!). This could allow increasing the salaries of the remaining work force. For instance, 50% of savings realised could go to that, and the rest (the other 50%) to increase O + M resources. Contributing to the same end, donors may have to show more willingness to pay for some of the O + M costs in their own projects, yet avoiding subsidising civil servants’ salaries in the long run. (Langseth, 1995)
Given that a high percentage of the payroll is in the lower job groups, annual savings in salaries from downsizing 25% of those employees (excluding teachers) could be substantial. Taking Kenya as an example, if the savings were split 50/50 and one half was earmarked for increases in salaries for the higher job groups, the 1990 salaries could have been increased by 36% for those groups (H and above) and the other half could have boosted O + M funds by 19%.
An alternative to putting savings realised back into salary increases could be to pay semi-annual bonuses based on performance as a means of introducing an incentive factor in the emolument structure of civil servants. (Perry et al, 1989; SES, 1989; Baker, 1990) Here, one has to be aware of risks of nepotism, tribalism, gender discrimination, sexual harassment and favouritism in the granting of these performance bonuses. Given national cultural differences, the authors have no golden rule to recommend to avoid these negative behaviours.
The dilemma that arises, though, is whether to start by using the savings to introduce these incentives for the higher job groups rather than for those in the lower job groups that remain after the trimming operation. Incentives for the former are important to retain good technical cadres when faced with the growing competition exerted by the private sector. (Holzer, 1991) Explicit guidelines will, therefore, be needed to keep the good and disincentivate poorly performing cadres in these higher job groups, encouraging the latter into early retirement. In this sense, pruning the less productive individuals in the higher job categories should actually be a parallel exercise to the main downsizing. (Anderson, 1996)
Any mix of the above options could become a donor precondition if donor funds are to be used for the downsizing.
Donor contributions to golden handshake schemes: An endowment, a grant or a loan?
Since donors are increasingly showing willingness to co-finance civil service downsizing (World Bank, 1991; Langseth, 1995), it is an attractive idea to explore the possibility of achieving this goal by setting up an endowment. The interest earned from the same could pay for any of the different modalities of the golden handshake we explore in this article. These include a salary support after dismissal and a single lump sum payment. This ‘economic jumpstart’ approach supports workers while they readjust to their new occupations thus relieving fears and potential adverse political pressures.
The first strong argument we want to make here again is that downsizing is in the interest of long-term development. This is already understood by many donors. We think civil service reform has equal or more merit for donors than any technical project they support in, for instance, agriculture or health.
We think it has a potentially higher return on investments than other sectoral projects. This because it creates an overall enabling public sector environment with more needed O + M resources being made available to technical cadres to eventually be more effective in their work. We, therefore, contend that donors should look more closely to funding such downsizing endeavours, either in the form of a grant, a loan or perhaps an endowment. We remain convinced that civil service reform is not a riskier venture for donors than any sectoral project they currently fund, especially when looked at in the long run.
Given the magnitude of the funding required for significantly downsizing bureaucracies, such funding will probably have to be pooled from various donors.
Why were we first attracted to the idea of setting up an endowment fund rather than for donors to give a grant or to arrange for a loan?
In the case of an endowment fund set up through a multi-donor commitment, the interest generated would pay for a temporary salary support of the dismissed civil servants every year for a fixed period after their dismissal (see illustrative example below).
The endowment, we posit, could be invested in the developed countries and be managed by a professional fund manager (outside government). Only the dividends (interests) would then be brought into the country in question. We predict that investing the endowment in local companies would have an inflationary effect and that donors would probably be more reluctant to follow such a course of action.
Conversely, an exhaustible fund could be set up that each year would disburse money partly in the form of interest accrued plus a part of the principal, in a way that when the downsizing target is reached, the funds are all spent. A fund of the latter type would, of course, be significantly lower.
An endowment fund could also be attractive to donors, because they could pull their principal out after the trimming operation is completed.
Lastly, we were attracted, because the endowment principal could be movable. Therefore, when 75 or X% of the goal of trimming the bureaucracy had been achieved in a given participating country, the endowment could be used to successively help other countries do the same. In other words, one could talk of a sort of a movable global endowment. As said, the endowment idea has to be weighed against a lump sum donor(s) contribution or a soft loan disbursed over 3 years.
The Kenya example
To illustrate the size of a hypothetical endowment we did some calculations in a way that, at prevailing interest rates in the developed countries in 1993, the endowment would have yielded enough cash to give an economic jumpstart to 60,000 Kenyan civil servants, excluding teachers. Leaving the service over a 3 year period would have roughly decreased the number of employees of job groups A-G by 25% by the end of 1995. The results we obtained were, in a way, not so encouraging.
Investing the fund in the economies of the developed countries at 7% interest in dollar terms, with no capital gains, the endowment would have to be of a magnitude of around US$ 408 million, roughly yielding US$ 28 million interest per year, enough for 20,000 people’s golden handshakes.
At an 8% dividend rate, the endowment for 20,000 staff per year would amount to around U$ 357 million yielding the same US$ 28 million needed per year.
If the endowment additionally had capital gains of 5% per annum, the endowment would go down to around U$ 238 million, a figure that seems more manageable. Note that the latter level of yields -in the 10-12%/year range- seems entirely possible in mid 1997 as this article is being finalised.
The modality of how to use these benefits to give leavers a golden handshake can vary. We chose an option that included: full salary for 6 months; a balloon payment equivalent to 18 months of full salary on the 7th month and thereafter, half salary for 2 more years. For this, the endowment would have to be invested several months before starting the programme to accrue enough dividends to pay the balloon.
In total, this is equivalent to 36 months average salary paid over 30 months.
As can be seen from this example in Kenya, donor inputs of this magnitude are rarely heard of, even if one could sell the idea of the movable, global endowment. Therefore, grants or loans (or mixtures thereof) of the order of U$ 28 million per year, over three years, would be needed for a country like Kenya to carry out a sizeable enough downsizing of its bureaucracy over three years using the suggested golden handshake modality.
The options here presented have advantages and disadvantages. Together with others, we do not claim to have final answers to the many questions raised by them. (Falcon, 1991) But all the options we do suggest have, we think, some merit to be explored in each specific national context for possible adoption and adaptation.
The golden handshake: A grant or a loan to departing civil servants?
What follows, explores available options for golden handshake payment modalities. This includes those in which a part of the package paid to compensate leavers is to be repaid by them.
The question thus is whether payments to them should be in the form of a grant or a loan (the latter soft or hard?) or a combination of both.
Moreover, one can ask whether the payment should be all in one lump sum paid upfront, or should the golden handshake rather pay the equivalent of the leavers’ monthly salary, or part thereof, for a given number of months. Alternatively, a third option may be more desirable: monthly salaries could be continued with a balloon payment after 6, 12 or 18 months of leaving the service. The monthly emolument can then be cut a significant percentage after the balloon payment. This latter option is based on the assumption that, after some time, the leaver may have a better idea where and how to invest his/her money.
The system could also be set up on a decreasing sliding scale: for the first 3 months, the employee would get 100% of his/her salary; for the 4th to the 6th month 75%; for the 7th to the 9th month 66%; for the 10th to the 12th 50%, and from the second year on, 33%. The progression chosen could also be slower, perhaps over 18 or 24 months. This monthly emolument would go up proportionately if remaining civil service employees got a salary rise.
Should it be decided that the balloon payment is to be a loan and not a grant, the loan could be managed in a “Grameen Bank-like” fashion requiring groups of 3 or more civil servants to take joint responsibility for repaying. The loan could be disbursed in two installments first to establish a repayment capability. (*)
(*): “Grameen Bank-like” is here used as an adjective denoting mostly the spreading of risk as an alternative to collateral required by the lender.
Several further questions arise at this point: Would a grace period be necessary? If so, for how long? Would the investment projects the leavers have in mind be reviewed before getting their money, or would the loans be given with ‘no questions asked’? One certainly does not want to create a new bureaucracy to manage the downsizing exercise.
Could private institutions manage the loan under contract with the government or with the donors -for example, an NGO already in the business of granting small credits? Would direct loans actually be necessary at all, or is it better, and cheaper, for the government to just establish a collateral with donor funds and channel the leavers -as borrowers- to a state or private bank?
We are keenly aware that a loan system, as opposed to a grant system for the upfront or balloon payment, may simply not be attractive enough to the majority of leavers who have no entrepreneurial inclinations.
If an upfront grant were given, would the grant be related to the leaver’s salary? Higher paid servants would thus get a bigger grant raising issues of equity. This risk would be reduced if the balloon payment were a loan.
Finally -cultural biases aside, and discounting probable adverse reactions to such a proposition- would it be viable to give part of the upfront lump sum or balloon payment to the wives of male leavers?
As said earlier, this article does not intend to have answers to all these questions. But the questions stimulate the type of creative thinking that we contend is needed to come up with the best mix of options for a variety of different possible contexts.
To give incentives or to dismiss
In general, launching a survey on a sample of the targeted group of civil servants -asking them what inducement(s) they would deem needed to make them consider resigning- would be of great help to better set the terms of a golden handshake package. The results could not be taken fully on their face value, but would help planners design something that approaches the preferred financial options as voiced by the target group.
Should the golden handshake or economic jumpstart be given as an option or should it dismiss civil servants regardless of their wishing to join voluntarily -or should it use a combination of both approaches? We seem to agree that it is desirable to start it as a voluntary operation and then -when interest wanes- add a forced dismissal component. Putting up a credible threat or broad hints of dismissal could improve performance of civil servants leaving behind what in Northamerican English has metaphorically been called a “lean and mean” civil service. (Perry, 1991) The procedures for disciplinary cases to become first candidates for the dismissal operation should be worked out in advance and used. Perhaps they could receive fewer benefits than civil servants in good standing. This would especially apply for staff having been involved in fraud, theft or misappropriation of funds or with a history of alcohol abuse or absenteeism. Dismissal operations should ideally target the least productive employees of any level or job group. A ‘last in-first out’ policy may otherwise have to be adopted for such an operation.
Beware that a ‘witch hunt’ may develop from this dismissal operation, including one with ethnic, gender or other discriminatory overtones. A special appeals committee may be needed to arbitrate in such cases.
When launching an incentive-based golden handshake operation, a deadline would be required. Civil servants should know dismissal will be a certainty at the end of a set deadline, if the trimming exercise has not reached the expected targets.
A good system of publicity around the downsizing exercise, openly explaining all options, will decrease levels of apprehension among civil servants. Essential service providers in the organisation, even at lower grades, who apply for the golden handshake, may be granted approval only after a suitable replacement has been found. The gender bias issue needs very special attention in downsizing. It even is conceivable to slant the bias in favour of more often keeping women in lower job groups.
Other implementation issues: Alternatives on how to set up the payment system for the golden handshake:
It is interesting to note here that, in Ghana, World Food Programme commodities were used for quite a few months as part of the severance package given to civil servants being dismissed. In Uganda, UNDP funds were used to downsize the army. (Langseth, 1995)
Details on whether the golden handshake will be a one time lump sum payment only or a percentage of the employee’s salary to be paid over X number of months, and details on whether to include a balloon payment at all will be highly country-specific. Whether the government will set up the system, as opposed to a bank, will also vary.
A couple of additional questions come up at this point of the analysis: How can donors, eventually footing part of the bill, be satisfied that the trimming is actually happening? What proof could they insist on getting about the staff reduction process’s progress? Among other things, they could: look at the government’s monthly payroll and disburse their funds for the golden handshake instalments, only as they have proof of reductions in the civil workforce. have civil servants get a severance certificate and draw all their golden handshake benefits from an entity that has donor control, for example a savings account in a commercial bank. (*)
(*): The scheme would thus effectively be financially administered outside the government.
Place advertisements in the local newspapers announcing the option for lower grade civil servants to leave their employment and become eligible for a package of benefits. Interested civil servants would then resign, get their severance certificates and go with them to designated banks and begin drawing funds as per the advertised package. The reaction of civil servants would then be observed for a few months. If there are very few takers, a new advertisement in the papers would present an improved package deal and the reaction would be observed again. The takers of the first package would automatically become eligible for an upgrade to bring them up to par with the new package. This can be repeated 2-3 times, if necessary, until the package is attractive enough to reach the desired rate of progression and percentage of reduction in the lower grade staff. Should the package elicit a response bigger than expected -beyond what had been planned for financially- it would be made clear to civil servants that the system will either work on the basis of first-come-first-served, or a lottery system would have to be set up.
One should be very clear that the risk of losing the good people first is unavoidable. But this may not matter. The preferred option would still be a minimal acceptable incentives package followed by mandatory severance if the voluntary downsizing does not result in the expected reduction.
Social security (retirement) benefits would still accrue to the leavers at retirement age according to the prevailing regulations in the country. They are deemed to be their rights. This should be made clear to civil servants as part of their severance package briefing.
Pilot operations research projects are advisable in adopting any of these approaches. One could start by dismissing some 500-1000 civil servants under different dismissal schemes, with tracer studies over two years. Such pilot schemes have the advantage of getting an overly apprehensive government interested, allaying fears about the political costs and helping identify difficulties in the implementation of a large programme later.
Finally, we totally discounted the feasibility of any plan that would offer any kind of government bonds or papers (as opposed to cash) as part of the payments for the downsizing operation.
To phase or not to phase:
The possibility for a staged approach to launch a golden handshake scheme also exists. One could:
In stage 1
Go to a 4-day working week giving one day off to civil servants choosing to leave and adding 1 more working hour to each of the 4 remaining working days. These servants would then be expected to identify or start some economic activity during this transition period, in anticipation of severance. (*)
In stage 2
Use a hypothetical goal to discuss the number of civil servants one wishes to trim, for instance taken from the experience of another developing country that has achieved better P/O + M/CI ratios or from a Delphi group discussion (as described earlier) and proceed with the downsizing exercise along these lines.
In stage 3
Fine-tune the above goal using the country’s own experience gained during downsizing during Stage 2, or some more sophisticated method(s) can also be tried. (**)
(*): As far as we know, this has not been tried anywhere, but it certainly has merits to see whether the efficiency of the bureaucracy would suffer or improve.
(**): For example, zero-based budgeting that looks at workloads and performance standards. (Glen, 1990; Fox, 1991)
The point we emphasise here is that there is no excuse for delaying the launching of the downsizing exercise with the pretext that the ideal mix of P/O + M/CI is still being studied!
How to redeploy public servants to the private sector?
To train or not to train:
It would be desirable to know what assets, connections, skills, and therefore, prospects lower echelon civil servants have. A survey could be designed and carried out to this effect prior to starting downsizing operations. This would give a better idea of what civil servants intend to do if and when they leave. Quite a bit could be learned that could influence the final characteristics of the golden handshake package.
If the leavers are to be absorbed by the existing market economy enterprise sector in large enough numbers, these severed civil servants should have the option to be trained prior to (or at the time of) taking up their new activities in that sector. The question is, at whose expense: the new employers’, the government’s or the donors’? Training may or may not thus become a part of a golden handshake scheme and a small pilot training fund could be set up for this purpose. (Paul, 1983)
On the other hand, it is not a bad idea to earmark part of the savings in salaries achieved by downsizing to be used to upgrade the technical skills of the remaining staff, i.e. giving in-service training to those who stay.
Absorption in the private sector:
The best way to increase the absorptive capacity of the private sector is through private sector promotion policies and some deregulation. Such measures should go along with the civil service reform.
In any case, the rate of downsizing should be kept at a reasonable fraction of the rate of new workers entering the job market. This is not usually an insurmountable problem. In Kenya, for example, even 40,000 dismissals per year (double the figure that was proposed earlier) would only have been equal to 10% of the growth of the labour force in 1993. An overall policy environment for private sector growth would make a far larger than 10% difference in the rate of job creation.
The capabilities of the private sector to absorb this labour are certainly not unlimited under present circumstances in most developing countries. The absorptive capacity of this sector needs to be estimated more precisely during the preparatory work to be carried out for the downsizing. Macro and microeconomic factors will have to be taken into consideration in the latter exercise.
One of the potential areas of economic activity these redundant civil servants can be geared towards, particularly in Africa, is the small-scale agricultural sector that is currently applying more modern technologies.
The Kenyan example can illustrate the magnitude of the absorption capacity required by the private sector. The increase that would have been needed in the country’s economic expansion to absorb roughly 20,000 former civil servants per year would not have been that great in the early nineties: it would have meant increasing the Kenyan labour force by approximately 5% every year for 3 years.
Otherwise, should incentives be considered for potential employers of former civil servants? Foreign exchange retention propositions for exporters, as well as tax break incentives considered for firms that hire dismissed civil servants are potentially very dangerous. We predict companies will tend to dismiss their old staff to hire these civil servants to become eligible for these economic incentives.
Finally, also note that the concepts here discussed for trimming the bureaucracy can also be applied to a downsizing exercise in parastatal organisations.
Conclusions:
A major concern to keep in mind is that probably no more than 25% of the lower-echelon government employees are entrepreneurial enough to start up a business if dismissed and given some lump sum or balloon payment. Of particular concern is whether these are the 25% that would leave if incentives for resigning are given, thus leaving behind a conceivably less productive staff.
This concern brings us back to the central question whether one should only strive to cut the bureaucracy or, at the same time, strive to launch the leavers into (independent) economic activities in the private sector. Of course, these economic activities include wage employment in the private sector, and a good number of leavers are expected to find private sector employment.
Civil service reform is not a cure-all remedy to all developing countries’ ills. Downsizing does lower the complement and thus lowers public debt. It does have the potential to improve the efficiency of the bureaucracy through higher salaries and through the increased availability of complementary O + M funds. On the other hand, the economic jumpstart measures proposed can cushion -but only cushion- the economic impact on the leavers when on their way out of government employment. (Nunberg, 1990)
Entrepreneurial civil servants, those who perhaps already have outside economic activities, will most probably look forward to get help to make a smooth full transition to the private sector. Less or non-entrepreneurial civil servants will be more reluctant to move, but cash is still the only incentive that will finally make them leave.
Six to eight months after the proposed balloon payment, a loan facility may be opened specifically for those 20-25% who we expect will make it in business ventures after leaving. This entrepreneurs’ fund, with donor start-up capital, could be administered by an NGO. The monthly golden handshake emolument, if still received, can be used as partial collateral by potential borrowers.
It is not to be forgotten that downsizing by itself does bring about significant savings, such as less government housing subsidies, less office supplies, less transportation and utilities costs, and many other savings. But massive dismissals -which is the shortest route to trimming- are, we said, unacceptable, not only politically, but also in terms of social justice and fairness.
We are aware that we have here mostly made a number of attractive, but untried suggestions that bring in some fresh thinking into the increasingly urgent task (and challenge) of trimming bureaucracies in developing countries. Given the implications it has for the economy of a country as a whole, downsizing the civil service is at least as appealing a ‘development’ project as traditional, sectoral development projects.
Every country is different and this article has only suggested paths or avenues worth exploring when starting to design civil service trimming operations. Important questions have been brought up, for which national policy makers will have to find answers. We have only shown some additional ways to think about what needs to be done.
The political constraints to downsizing will still, for a long time to come, be the core determinant to get downsizing processes even seriously explored, let alone started. But downsizing is overdue in many places and ours is a contribution to demystify the process with a hope that this will remove some of these constraints and unjustified fears. We think that bold donors can help catalyse the process.
References:
Adamolekum, L. (1989). ‘Issues in development management in Sub-Saharan Africa’, EDI Policy Seminar Report, World Bank, Washington, DC.
Anderson, D. (1996). ‘The true cost of early retirement schemes’, presented at the Seminar on Staff Reductions as a Consequence of Modernisation and Restructuring of the Administration, Sofia, Bulgaria, May.
Baker, J. (1990). ‘Improving public sector performance: A world view’, Canberra Public Service Commission.
Balasz, I. (1996).’Staff reductions in Hungarian public administration and the experiences of implementation’, presented at the Seminar on Staff Reductions as a Consequence of Modernisation and Restructuring of the Administration, Sofia, Bulgaria, May.
Ban, C. et al. (1992). ‘Productivity and the personnel process’ in Public Productivity Handbook, Mark Holzer, Ed., Jossey-Bass, San Francisco.
Binder, AS. Ed. (1990). Paying for Productivity, The Brookings Institution, Washington, DC., de Merode, L. (1991). ‘Civil service pay and employment reform in Africa: Selected implementation experiences’, World Bank, Africa Technical Department, Washington, DC., June.
Falcon, D. (1991). ‘Life after Thatcher: Trends in UK public administration and lessons for developing countries’, Notes on lecture sponsored by the Public Sector Management and Private Sector Development Division, World Bank, Washington, DC., March.
Fallon, P. and Pereira da Silva, L. (1991). ‘Public pay and employment issues in Mozambique’, World Bank, Washington DC.
Fox, C. (1991). ‘Employee performance appraisal: the keystone made of clay’, in Public Personnel Management: Current Concerns, Future Challenges, Carolyn Ban and Norma Riccucci, Eds., Longman, New York, pp. 58-72.
Fry, G. et al. (1988). ‘Symposium on improving management in government’, Public Administration (UK), 66 (Winter), pp. 429
Glen, R. (1990). ‘Performance appraisal: An unnerving yet useful process’, Public Personnel Management, 19:1 (Spring), pp. 1-10.
Holzer, M. (1991). ‘Attracting the best and brightest to government service’, in Public Personnel Management: Current Concerns, Future Challenges, Carolyn Ban and Norma Riccucci, Eds., Longman, New York, pp. 3-16.
Landau, M. (1969). ‘Redundancy, rationality and the problem of duplication and overlap’, Public Administration Review, 29:4, (July-August), pp. 346-358.
Langseth, P. (1995). ‘Civil service reform in Uganda: Lessons learned’, Public Administration and Development, 15, pp. 365
Nunberg, B. (1989). ‘Public sector pay and employment reform: A review of World Bank experience’, World Bank Discussion Paper 68, World Bank, Washington DC.
Nunberg, B. (1990). ‘Bolivia: A review of public pay and employment issues’, Country Economics Department, World Bank, Washington DC.
Nunberg, B. and Nellis, J. (1990). ‘Civil service reform and the World Bank’, PRE Working Paper 422, Country Economics Department, World Bank, Washington DC.
Ozgediz, S. (1983). ‘Managing the public service in developing countries: Issues and prospects’, World Bank Staff Working Paper No. 583, World Bank, Washington, DC.
Panov, O. (1996). ‘Staff reductions as an element of the administrative reform in Bulgaria’, presented at the Seminar on Staff Reductions as a Consequence of Modernisation and Restructuring of the Administration, Sofia, Bulgaria, May.
Paul, S. (1983). ‘Training for public administration and management in developing countries: A review’, World Bank Staff Working Paper No. 584, World Bank, Washington, DC.
Perry, JL. et al. (1989). ‘Federal merit pay, Round II: An analysis of the performance management and recognition system’, Public Administration Review (January-February), pp. 29-37.
Perry, JL. (1991). ‘Linking pay to performance: The controversy continues’, in Public Personnel Management: Current Concerns, Future Challenges, Carolyn Ban and Norma Riccucci, Eds., Longman, New York, pp. 73-86.
Prokopenko, J. (1989). ‘Improving public service productivity: Problems and challenges’, International Labour Office, Management Development Branch, Geneva.
Quah, JST. (1986). ‘Toward productivity and excellence: A comparative analysis of the public personnel systems in the ASEAN countries’, in Delivery of Public Services in Asian Countries: Cases in Development Administration, Suchitra Punyaratabandhu et al. Eds., Thammassat University Press, Bangkok, pp.238-272.
SES, (1989). ‘A new approach to pay and incentives’, Executive Bulletin (New South Wales), 1:3, (summer), p.3.
Synnerstrom, S. (1995). ‘Workshop examines staff reduction in the public service’, PMF, 1:4, p.5.
Synnerstrom, S. (1996). Personal communication.
Toole, BJ. (1989). ‘The next steps and control of the civil service: A historical perspective’, Public Policy and Administration, 4:1, (Spring), pp. 41-52.
Wallis, M. (1989). Bureaucracy: Its role in Third World Development, Macmillan, London.
World Bank, (1991). ‘The reform of public sector management: Lessons from experience’, Policy and Research Series 18, Country Economics Department, World Bank, Washington, DC.
World Bank, (1997). ‘Getting it right the first time: Research project cautions approaches to downsizing’, World bank News, XVI : 9, March 6, pp. 1-2.
Zussman,D (1996). ‘Downsizing as a means to an end: Lessons learned’, presented at the Seminar on Staff Reductions as a Consequence of Modernisation and Restructuring of the Administration, Sofia, Bulgaria, May.
Claudio Schuftan
Saigon,Vietnam.