Why leadership in the public sector is harder to find--but more important

Occasionally I post about things that are related to recruitment and assessment, but not focused exclusively on them.  This is one of those times.

I have the following quote from Valve Software's New Employee Handbook (a fascinating document) posted on my office door:

"Hiring well is the most important thing in the universe.
Nothing else comes close. It’s more important than breathing.
So when you’re working on hiring—participating in
an interview loop or innovating in the general area of
recruiting—everything else you could be doing is stupid
and should be ignored!"

The older I get, the more I wonder whether I should cross out "hiring" and write in "leadership".  I just can't bring myself to do it because they're both so darn important.

But this post will be about leadership.  Specifically, leadership in the public (i.e., government) sector.  More specifically, lack of leadership and what to do about it.  I don't pretend that leaders in the private sector are uniformly outstanding, but public sector is what I'm most familiar with.

First things first: in some important ways, leadership in the public sector (PS) is different from the private sector.  Not night-and-day I grant you, but there are some relatively unique boundary conditions that apply, namely:

- Not only are PS leaders bound by normal organizational policies and procedures, they labor under an additional layer of laws and rules, whether federal, state, or local.  Unlike policies and procedures, they cannot be easily changed--in fact in many cases this requires moving heaven and earth.  As one (very important) example, typically there are laws/rules about how you can hire someone.  Many of these laws and rules were created 50+ years ago in reaction to spoils systems and haven't been seriously evaluated since.

- Many PS employees have civil service protections.  This isn't a bad thing, but it means moving employee that are bad fits (either over or out) is difficult.  This greatly inhibits your talent mobility strategy.

- In the PS, leadership is often treated as an afterthought, rather than the linchpin upon which organizational success relies.  The assumption seems to be that the organizational systems and processes are so strong that it almost doesn't matter who's in charge.  This means things like leadership development and training are half-hearted.

These conditions combine with several other factors to result in true leadership being relatively rare in the PS:

- Failure is invisible.  There is very little measurement of leadership success and very little transparency and accountability, absent a media storm.

- There are fewer people in management positions that have the important leadership competencies.  Things like listening ability, strategic planning ability, and emotional intelligence.  Instead they are usually chosen based on technical ability and without the benefit of rigorous assessment results.

- There is a lack of understanding of leadership.  This stems from the lack of attention paid to it as a serious discipline; without operational definitions of leadership, there is no measurement and no accountability.

- An unwillingness to treat leadership seriously.  For whatever reasons--politics, lack of motivation, entrenched cultures--leadership is relegated to second-class status when it comes to analyzing department/agency success.  Focus tends to rest on line-level employees, technology, and unions--and only on top-level leaders when there is a phenomenally bad outcome.

So why is leadership more important in the PS than the private sector?

- Governments regulate many aspects of our lives.  They're not making consumer products.  Leaders in PS organizations have purview over things like public safety, the environment, education, housing, and taxes.  Things you literally touch every day.

- There is less accountability, less transparency.  PS leaders often do not report to a board.  They don't have to produce annual reports that detail their successes and failures.  What they do is often mysterious, poorly defined, and rarely sees the light of day.

- PS leaders work for you.  Elected or not, their salary typically comes from taxpayers.  They ultimately report to the citizens.  This means you should care about what they're doing, and whether they are worthy stewards of your investment.

So what can be done?  Much like the answer to "how do we hire well?", the answers are known.  They're just not practiced very well:

1.  Publicly acknowledge the scope of the problem.  Like frogs in a pot, somehow we find ourselves in a situation where slowly over time the current situation is accepted as normal.  It's time to stop pretending that all PS Managers are leaders.  They're not.  And we must look in the mirror ourselves and acknowledge that we are likely part of the problem.

2.  Acknowledge the urgency to improve.  Stop pretending that leadership is a secondary concern.  Sub-par leadership has a negative impact on our lives every day.  Improving the quality of that leadership is one of the most critical things we can do as a society.

3.  Publicly commit to change, and actually follow through.  Specifically describe what you will change, and when, and provide regular status updates.

4.  Define leadership in measurable terms and behaviors.  Here's just a sample list of what real leaders  do (and not a particularly good one);

= continuously improve operations
= champion and reward innovation
= hold their people accountable for meeting SMART goals
= continually seek feedback and signs of their own success and failure
= create and sustain a culture that attracts high performers and dissuades poor fits
= make hiring and promoting the most qualified people THE most important part of their job

5.  Hire and promote those with leadership competencies, not the best technicians.  While knowledge of the work being performed is important, it is far from the most important competency.

6.  Make the topic of leadership a core activity for every management team.  Eliminate "information sharing" meetings and replace them with discussions on how to be better leaders.

7.  Set clear goals of leaders up front, and hold them accountable.  What does this mean?

= consequences for hiring poor fits
= consequences for poor morale on their team
= consequences for not setting and meeting SMART goals
= recognition for doing all of the above well

8.  Measure leadership success and make the results transparent.  Develop plans to address gaps and follow through.

9.  Instill a culture of boldness and innovation.  Banish fear, often borne of laboring under layers of red tape.  Encourage risk-taking, and learn from mistakes rather than punishing them.

10.  Relentlessly seek out and banish inefficiencies, especially related to the use of time.  Critically evaluate how email and meetings are used; establish rules regarding their use.

11.  Stop pretending that all of this applies only to first-line supervisors.  If anything, they're more important the higher you go in the organizational chart.

12.  When it comes to recruiting, stop focusing on low relative salaries, and capitalize on the enormous benefit of the PS an employer--namely the mission of public service.

13.  View leadership as a competency, not a position.  Leadership behaviors can be found everywhere in an organization--they should be recognized and promoted.


My intent here is not to be a downer, but to emphasize how much more focus needs to be placed on leadership in the public sector.  The current state of affairs is unacceptable.  And for those of us familiar with research and best practices in organizational behavior, it's painful.

So I apologize for the decidedly un-Thanksgivingy nature of this post.  But I am thankful for free speech and open minds.  Thanks for reading.

WRIPAC's 35 Anniversary Meeting: Something old, something new, something borrowed, something to do

Over the last two days I had the pleasure of attending the Western Region Intergovernmental Personnel Assessment Council's (WRIPAC) 35th anniversary meeting in San Francisco.  I thought I would share with you some of my observations, as it's been a while, unfortunately, since I attended one of their events.

For the uninitiated, WRIPAC was founded in 1979 and is one of several regional associations in the United States devoted to personnel assessment and selection in the public sector.  Other examples include PTC-NC, PTC-SC, PTC/MW, and IPAC.  I have been involved with several of these organizations over the years and they provide a tremendous benefit to members and attendees, from networking to best practice to the latest research.

WRIPAC serves public agencies in Arizona, California, Nevada, and Oregon.  They have a somewhat unique membership model in that there is no membership fee and the meetings are free, but in order to become a member you have to attend two consecutive meetings.  They maintain their energy in large part through a commitment to their committees and providing excellent training opportunities.

So, about the 35th anniversary WRIPAC meeting, hosted by the City and County of San Francisco:

Something old:  One of my observations over the day-and-a-half meeting was the remarkable number of issues that seem to be the same, year after year.  For example:

* how to effectively screen large numbers of applicants

* how to handle increased workload with a reduced number of HR staff

* how to convince supervisors and managers to use valid assessment methods

* which assessment vendors provide the best products

* how to design selection systems that treat candidates fairly but don't unduly burden the agency

We also were treated to a retrospective by three previous WRIPAC presidents, which highlighted some really stark ways the workforce and the assessment environment has changed over the years.  This included everything from sexual harassment in the workplace being commonplace to agencies trying to figure out how to comply with the new (at the time) Uniform Guidelines.

Something new:  Dr. Scott Highhouse from Bowling Green State University presented results from several fascinating studies.

The first looked at "puzzle interview questions"--those bizarre questions like "Why are manhole covers round?" made famous by companies like Microsoft and Google.  Using a sample of participants from Amazon's Mechanical Turk (the first time I've heard this done for a psychological study), he was particularly interested in whether individual "dark side" traits such as machiavellianism and narcissism might explain why certain people prefer these types of questions.

Results?  First, men were more likely to endorse these types of items.  Why?  Well it may have to do with the second finding that respondents higher in narcissism and sadism were more likely to endorse these types of questions.  From what I can tell from my massive (ten minute) search of the internet, men are more likely to display both narcissism and sadism.  Maybe more importantly, the common denominator seemed to be callousness, as in being insensitive or cruel toward others.

So what does this mean?  Well I had two thoughts: one, if you like to ask these types of questions you might ask yourself WHY (because there's no evidence I know of to support using them).  Second, if you work with supervisors who show high levels of callousness, they might need additional support/nudging to use appropriate interview questions.  Looking forward to these results getting published.

The second study Dr. Highhouse described looked at worldviews and whether they impact beliefs about the usefulness of testing--in particular cognitive ability and personality tests.  This line of research basically is trying to discover why people refuse to entertain or use techniques that have proven to work (there's a separate line of study about doctors who refuse to use decision aids--it's all rather disturbing).

Anyway, in this study, also using participants from Mechanical Turk, the researchers found that individuals that had strong beliefs in free will (i.e., people have control of their choices and should take personal responsibility) were more open to using conscientiousness tests, and people with strong beliefs in scientific determinism (i.e., behavior stems from genetics and the environment) were more open to using cognitive ability tests.  This adds valuable insight to why certain supervisors may be resistant to using assessment methods that have a proven track record--they're not simply being illogical but it may be based on their fundamental beliefs about human behavior.

The last study he talked about looked at whether presenting people with evidence of test validity would change their related world views.  You wouldn't expect strong effects, but they did find a change.  Implication?  More support for educating test users about the strengths and weaknesses of various assessments--something we do routinely but for good reason!

Last but not least, Dr. Highhouse introduced a journal that will hopefully be coming out next year, titled Journal of Personnel Assessment and Decisions, that will be sponsored by IPAC and Bowling Green State University, that aside from the excellent subject matter, will have two key features: it will be free and open to everyone.  I'm very excited about this, and long-time readers will know I've railed for years about how difficult it is for people to access high-quality research.

Something borrowed:  one of the big benefits of being involved with professional organizations--particularly ones like WRIPAC that are practitioner-focused--is it gives you access to others facing similar challenges that have come up with great solutions.  There was a lot of information shared at the roundtable about a wide variety of topics, including:

- reasonable accommodation during the testing process (e.g., armless chairs for obese individuals)

- how to use automated pre-screening to narrow candidate pools

- how agencies will adjust to the new requirements in California related to criminal background information (i..e, "ban the box" and other similar legislation)

- how to efficiently assess out-of-state applicants (e.g., video interviewing, remote proctors)

- how and when to verify a drivers license if required for the job

- how to effectively use 360-degree evaluations

- how MQs (the subject of June's free PTC-NC meeting) should best be placed in the selection process

- cloud vs. on-premise IT solutions

- centralization vs. decentralization of HR functions

- use of item banks (e.g., WRIB, CODESP)

In addition, there was an excellent entire afternoon session devoted to succession and leadership planning that featured three speakers describing the outstanding programs they've managed to administer at their agencies.  I took a ton of information away from these and it came at exactly the right time as we're looking at implementing these exact same programs.

Something to do:  One of the main things I took away from this conference is how important it is to maintain your participation in professional associations.  It's so easy to get sucked into your daily fires and forgot how valuable it is, both personally and professionally, to meet with others and tackle our shared challenges.  I plan on sharing what I learned back in the office and upping my expectation that as HR professionals we need to be active in our professional community.  I encourage you to do the same!

MQs: An idea whose time has passed?

For better or worse, I've spent nearly my entire career working under merit systems.  For the uninitiated, these systems were created many years ago to combat employment decisions based on favoritism, familial relation, or other similarly non-job related factors.  For example, California's civil service system was originally created in 1913 (and strengthened in 1934) to combat the "spoils" system, whereby hiring and promotion was too often based on political affiliation and patronage.

Part of most merit systems is the idea of minimum qualifications, or MQs.  Ideally, MQs are true minimum amount of experience and/or education (along with any licenses/certifications) required for a job.  They set the requirement to participate in civil service exams, and scale up depending on the level (or "classification").  For an entry-level attorney, for example, one would need to have a Bar license.  For a journey-level attorney, you might be required to have several years of experience before being allowed to examine and be appointed.  The idea is that MQs force hiring and promotion decisions to be based on job-related qualifications rather than who you know or what political party you belong to.  Makes sense, right?

But recently, I've had the opportunity to be involved in a task force looking at minimum qualifications and it spurred a lot of discussion and thought.  I'd like to spend just a moment digging into the concept a bit more and asking: are they still the right approach?

This task force was formed because of a recent control agency decision that places increased importance on applicants meeting MQs and reduces the ability of employees to obtain positions by simply transferring from one classification to another based on similarity in level, salary, etc.  Because this will result in fewer options for employees--and hiring supervisors--the discussion around this decision has been rigorous, and at times heated, but without a doubt intellectually stimulating.

As part of my participation in this task force, I reached out to my colleagues in IPAC for their thoughts, and got a ton of thoughtful responses.  While there were arguments for and against MQs, the overall sense seemed to be that they are a necessary evil.  Perhaps most importantly, though, I was reminded how important they are and thus the amount of attention that should be paid while establishing them.

So where does this lead me?  To play my hand, over time I've become less and less of a fan of MQs, and my participation on this task force has cemented some of the reasons why, however well intentioned:

- They are overly rigid and inflexible.  If an MQ states you must have 2 years as an Underwater Basketweaver, it doesn't matter than you have 1 year and 11 months and you just attended the Basketweaver Olympics, sorry, you don't qualify to test for the next level.

- They are often difficult to apply, resulting in inconsistencies.  What exactly is a four-year degree in "Accounting"?  What is "clerical" work?  If someone worked overtime, does that count as additional experience?  How shall we consider education from other countries?  And what about fake degrees and candidates who, shall  we say, elaborate their experience?

- They serve as barriers to talented individuals.  This results in fewer opportunities for people as well as a smaller talent pool for supervisors to draw from (ironically actually cannibalizing the very concept of the merit system).

- They serve as barriers to groups that have a history of discrimination, such as women and ethnic minorities.  Take a look at any census study of education, for example, and look at the graduation rates of different groups.  Implication?  Any job requiring a college degree has discrimination built into the selection process.

- Most were likely not developed as rigorously as they should have been.  Like any other selection mechanism, MQs are subject to laws and rules (e.g., the Civil Rights Act and the Uniform Guidelines in the U.S.) that require them to be based on job analytic information and set based on data, not hunches or guesses.

- Without a process to update them quickly, they rapidly become outdated, becoming less and less relevant.  Many classification in the California state system, for example, haven't been effectively updated in thirty years (or longer).  This becomes particularly painful in jobs like IT, where educational paths and terminology change constantly.

- They require an enormous amount of resources to administer.  At some point someone, somewhere, needs to validate that the applicant has the qualifications required to take the exam.  You can imagine what this looks like for an exam involving hundreds (sometimes thousands) of applicants--and the costs associated with this work.

- From an assessment perspective, MQs are a very blunt instrument--and not a particularly good one at that.  As we know, experience and education are poor predictors of job performance.  Experience predicts best at low levels but quickly becomes irrelevant.  Education typically shows very small correlations with performance.  As anyone that has experience hiring knows, a college degree doth not an outstanding employee make.  So basically what you're doing is front-loading your "select out" decisions with a tool that has very low validity.  Sound good?

- The ultimate result of all this is employers with MQs systems are often unable to attract, hire, and promote the most qualified candidates, while spending an enormous amount of time and energy administering a system that does little to identify top talent.  This becomes particularly problematic for public sector employers as defined benefit plans are reduced or eliminated and salaries fail to keep pace, resulting in these organizations becoming less and less attractive.

Recognizing these limitations, some merit systems (the State of Washington comes to mind) have recently moved away from MQs, instead evolving into things like desirable or preferred qualifications.  This presumably still outlines the approximate experience and education that should prepare someone for the position, but relies on other types of assessments to determine someone's true qualifications, abilities, and competitiveness.  I like this idea in concept as long as an effective system is put in place to deal with the likely resulting increase in applications to sift through.

The private sector, of course, does not operate under merit system rules, and have had to deal with the challenges--as well as reaping the benefits--associated with of a lack of rigid MQs.  They do this through increased use of technology and, frankly, significantly more expenditure on HR to support the recruitment and assessment function (particularly larger employers).  Of course some private sector employers adhere to strict MQs as a matter of course, and they would do well to think about the challenges I outlined above.

So where does this leave us?  Do MQs still serve a valuable perhaps?  Perhaps.  They hypothetically prevent more patronage, although anyone that has worked in a merit system can tell you this still happens.  Perhaps the strongest argument is that as more employers move to online training and experience measures (another example of an assessment device with little validity but quick, and cheap), MQs serve as a check, presumably helping to ensure that at least some of the folks that end up on employment lists are qualified.

But I would argue that any system that still employs MQs is basically fooling itself, doing little to control favoritism and ultimately contributing to the inability of hiring supervisors to get the best person--which is what a system of merit is ultimately about.  Particularly with what we know about the effectiveness of a properly administered system of internet testing, MQs are an antiquity, serving as a barrier to more job-related assessments and simply not worth the time we spend on them.  If we don't reform these systems in some way to modernize the selection process, we will wake up some day and wonder why fewer and fewer people are applying for jobs in the public sector, and why the candidate pools seem less and less qualified.  That day may already be here and we just haven't realized it.

R&A Software Failures Hurt Taxpayers, Too

We tend to think of successes and failures of applicant tracking systems and other recruitment- and assessment-related technologies as impacting businesses--much of what's written is about large organizations such as Microsoft and Google and what software they decide to adopt.

But public sector organizations are using these technologies as well. And when they fail, it hurts not only the organization but taxpayers as well.

Case in point: the state of Washington recently decided to abandon their efforts to implement SAP E-Recruiting after nearly three years and millions of dollars. The state will now go with a hosted solution which is estimated to be $700-800,000 a year cheaper (and hopefully much easier) to maintain.

Having been ringside for some of this, I can tell you the problem was not with motivation or energy, or even IT knowledge. I suspect that a lion's share of the problem was related to the complexity of the program. This would match reports I've read that a significant number of organizations are moving away from single-vendor HR solutions and going with simpler, targeted products. It's also possible that businesses find it easier to implement these programs because resources (particularly internal experts) are easier to move around and buy-off is easier to obtain.

I wish them luck on their next purchase, and hope they do due diligence in their research (you can often find others who have had problems). Some type of audit may help them determine exactly what went wrong and how to prevent it the next time around. It's not just a matter of time, energy, and expense on the part of the organization, these failures impact applicants, hiring supervisors, HR staff, and ultimately taxpayers.