What I Would Do About the Dilemma of Discretion

As I wrote here whether the government is hiring public employees or companies, it faces what I call the “dilemma of discretion.” Allow public sector managers to hire and fire whoever they please, and the government runs the risk of having their brother-in-law – or the brother-in-law of a politico who is in a position to threaten them – hired, and good employees fired.  But bind those managers with all kinds of rules, to hire those who score highest on a civil service test and only fire an employee after a complicated series of steps, and you create a legalistic playground for those who seek to get paid to do a job without actually doing it.   The civil service system and bidding rules, by making personnel and contractor arrangements non-voluntary, eliminate reciprocity in the employment and contracting relationship. Once a test is passed or a contract is won, the employee or contractor owes as little as he can get away with. Management often seeks to do as little as possible for the employees in turn. The result is ongoing, petty conflict over rules, a poisonous work atmosphere, and low productivity and quality – the government is a lousy place to work, and many firms refuse to do business with it.

For public employees, the central issue is motivation, which I rate on five point scale:  “on a mission,” “self-directed professional,” “just a job,” “not my job,” and “beat the system.”  The civil service system, to the extent it works at all, measures the ability to do a job, not the inclination.  That can only be determined by trial and error, by hiring workers and then firing those who prove to be on the bottom end of the scale.  Prevented from doing so by civil service rules once a certain level of tenure is achieved, however, it is those at the bottom that the public sector tends to attract and retain.  And, given the way they are “managed,” working for a public agency is such a miserable experience that most public employees are knocked down one level of motivation for every decade on the job.  (When I was about to pushed down to “just a job,” I left.)  Others start out that low and move lower.  No wonder so many hope for an early retirement incentive and an early exit, above all other career goals.  Indeed, overtime, sick leave when not sick, and early retirement with a full pension, are the holy trinity of public service.

If quality public services are going to be produced, a way must be found to upgrade the staff at public agencies.  Shifting compensation from retirement to the working years, as proposed in my prior essay, would help.  More carefully screening applicants would help too.  Here the civil service system needs to be revitalized.  More and better tests need to be given more frequently.  I was told, for example, that the City of New York once gave a plumbing test that required applicants to actually do some plumbing, but replaced it with a written test to save money – and the quality of city plumbers declined.  Moreover, the limitation of testing needs to be acknowledged.  Tests can measure minimal competence, not motivation, and should not be the sole criteria for hiring.  A minimum score should be a requirement, but a higher score does not necessarily identify the best person for the job.

It is at the other end, however, that things need to change the most.  The complicated procedures that make it difficult to fire a public employee who cannot or will not do the job should be scrapped.  Managers should not have to continue to work for months, or years, in a tense environment with an employee they deem to be unsuitable.  If they are going to be held accountable for the quality of the work, and they should be, those managers should get to determine their own staff.  Moreover, the idea that it is even possible to quantify the quality of a person’s work, and document it to meet a legal requirement, is a myth.  A large share of public employees, including teachers, police officers, and maintenance workers, work independently.  Any evaluation of their performance is going to be subjective, and the cost of documenting poor performance is so high, and takes so long, that it is no wonder that most managers do not bother.   It is also no wonder that public employees are harassed over nitpicking and meaningless rules; those are the only criteria that they can be held accountable for. 

At issue, as usual, is a question of equity.  Every time a public employee decides to shop at a different store, or change the goods and services they buy, to get a better deal, someone somewhere loses their job and has to find another one to satisfy them.  Meanwhile, private-sector workers have to pay public sector workers up front in taxes, and then must take whatever they get.  If public service recipients are to have the same human rights that public employees take for granted when they purchase goods and services in the marketplace, the criteria for dismissal needs to be much lower than some kind of  “misconduct.”  It must be possible to ask a public employee to seek employment elsewhere simply because some other worker is likely to do a better job for the public.  That is, in itself, a low standards.  New employees are green, and perform poorly until the learning curve is climbed, one reason why even private companies, with no limits on who they can fire, worry about turnover.  If a manager believes an existing employee’s performance is so weak they would rather absorb that cost and take a chance on someone new, then someone new they should have.

On the other hand, just allowing public sector managers to fire whoever they please raises the other specter of the dilemma – cronyism.  While some line employees in the public sector are underpaid (in cash, not total compensation) relative to their private sector counterparts, virtually all managers are, and if you get what you pay for, it is fair to assume that the quality of management in public agencies is not high.  It is a legitimate fear that good employees might be removed for no other reason than failure to kowtow to an overbearing boss, or to make room for an employee who is no better, or perhaps worse, based on personal relationship.  Almost across the board, the public sector features tremendous age discrimination inequities – with the young as victims.  Faced with budget pressures and given the opportunity, however, it is possible that public agencies would engage in the type of age discrimination common in the private sector – removing an employee just because he or she is older and, therefore, more expensive.  Therefore, there needs to be some replacement for the existing procedures to assure managers act in the best interests of the public.

I propose to resolve the dilemma this way.

Once a year managers would be able to decide they no longer wish to retain one or more of their existing employees, for any reason, with no procedural impediment.  Those employees would be given an opportunity to convince some other manager to take them on – a teacher could seek employment in another public school, a police officer at another precinct, an analysts at another agency.  If they were not successful after, say, a month, such employees would receive three weeks of severance pay and health insurance for each year of prior employment, over and above any credited vacation and sick leave accrued, and even before the beginning of unemployment insurance.  This would be charged to the budget of the manager who chose not to retain the employee.  Thus, if a manager no longer wanted a worker with 17-years of experience, they would have to be prepared to offer nearly a year’s severance pay out of their budget, plus accrued time.  Procedural impediments would be replaced with a financial disincentive to replace staff unnecessarily, a financial disincentive that would also provide a financial benefit to those forced to transition into a new position or career.

Financial considerations would limit the extent to which any given manager, or managers in general, could replace employees.  Those making wholesale changes would have to ask for an additional budgetary allocation, or get their work done with fewer staff.   Those managers would draw the attention of top management, and would then be on the hook for the results of those changes, and could be dismissed themselves if they don’t pay off.  If the quality of service did not improve, or the cost decrease, then the manager would have merely cost the government a bundle in severance pay for no reason.  Of course, if an employee was essentially worthless or worse, as are a small share of public employees, there would be little lost by asking them to seek employment elsewhere.

The legislature could, as a compromise, apply the new rules to only to new hires those existing employees with less than 10 year’s experience who agree to accept it.  But in keeping with my principle that total compensation should be equalized between those hired at different times, rather than using savings on new hires to reward those cashing in moving out, those who were eligible to be dismissed without protracted procedures should by paid more – I’d say five to ten percent more, based on the share of existing public employees who could be removed and not replaced without being missed.

What is the risk in such a system?  Well, long time superb New York City school teacher and later author Frank McCourt wrote that early in his career in the New York City schools, he was fired by a principal simply because of a personality clash.  There is a risk that some good employees would be lost along with the bad ones, for reasons like that.  But how great is the risk?  I said long time New York City school teacher Frank McCourt, who was quickly hired by another principal who was more than willing to capitalize on the weaker principal’s mistake by improving his own staff.  Decent public employees not retained at their current position would, in theory, have a union advocating for them with other managers (yes, this would involve doing something for the dues, but with less procedural conflict, there would be fewer grievances).  Higher-level managers might also be their corner, since retaining them would save a big financial hit in the current fiscal year.  And, in some cases, if there is a personality conflict, it is better for everyone to move on rather than be locked in a multi-year tense work environment.

Might there be a managers cartel, in which a manger’s union convinces its members not to rehire workers dismissed by others, making the prior boss look bad if the employee’s performance improves?  That is theoretically possible.  In the case of teachers, however, note that I have proposed three competing public school districts.  If one were to dump solid teachers to hire (state-certified) cronies, it is likely another would hire the now available solid teachers and dump cronies, with possible consequences for their political overseers.

Now consider the benefit of the proposed system.  The “beat the system” group would be gone, and some in the “not my job” group might change their mind about the minimum level of effort they have to put in.  For everyone at higher levels of motivation, work would get better, because they wouldn’t be dragged down by their less motivated peers. 

Removing impediments to dismissal is far more likely to improve public services than merit pay, another proposal advanced by public sector critics.  Merit pay requires managers in public agencies (who are those who cannot get higher paying jobs in the private sector) to correctly evaluate the performance of every employee relative to every other employee (despite the fact that many of these employees work independently and such evaluations are inherently objective) and perfectly calibrate the amount that each will be paid accordingly.  Moreover, unless all of those decisions were accepted as fair by the vast majority of the employees, motivation and morale would decrease.  Even with the best of intentions, and that can hardly be assured, experience in public agencies tells me this is unlikely to happen.  Merit pay would require public sector managers to meet unrealistic expectations. 

But it is reasonable for such managers, when intending to dismiss employees in the bottom ten percent to actually dismiss those who are somewhere in the bottom 20 percent.  That is what I propose.

I expect to discuss possible responses to the dilemma of discretion in public contracting in another essay.