First, there are rumors, then comes the announcement: “Relieved of command due to a loss of confidence.” So, it went for several Air Force officers, who were removed from command in 2023. Military officers have often been regarded as the embodiment of exceptional leadership. Countless books, shows, and movies depict military leaders’ incredible critical thinking skills, ability to function under pressure, and capacity to build exceptional teams capable of extraordinary achievements. In reality, however, not all officers are great leaders. In an era where data is ubiquitous, is there data available to identify the qualities possessed by successful commanders among military leaders? Are military commander assessments based on past performance, future potential, an individual’s affect, or are they based on a complex set of pre-determined variables?
We argue that the Air Force should modernize its promotion system in order to field a force of impartially-selected, critically-assessed commanders for its next fight. As it stands now, the Air Force has not identified what commander qualities it wants, much less put those qualities into doctrine. The result has been a commander selection board that has had to rely on a simple decision mechanism from the cognitive school of decision strategy, with its accompanying biases and subjectivity, in choosing officers that are just “good enough.” Instead, the Air Force should borrow from Army and Navy programs to develop a commander assessment program, deliberately identify the key qualities it requires in its commanders, and take a rational choice school approach to select officer candidates with the highest potential for successful command. Modernization and change is vital; the status quo is not working.
The Status Quo
In 2023, the Air Force used a boarded promotion and commander selection process to select future leaders for the service. Although the procedures vary slightly based on career fields, the foundation remains relatively similar today. In this annual board, members are chosen to select the next iteration of commander candidates. The Secretary of the Air Force provides board members instructions identifying their authorities, requirements, and considerations in evaluating records. Specifically, the Secretary directs the boards to recommend the best, fully qualified officers for promotions and, later, commander candidacy. Records assessment is based on a whole person concept, with job performance being the most critical factor and the other seven factors, including leadership, being weighed equally. During the assessment, board members review officer selection records, which are promotion packages that include performance evaluations, military decorations, an officer selection brief, any adverse information, and any written letters to the board members.
Traditionally, one to three minutes are spent reviewing each package containing data points encompassing a member’s career. Once packages are graded, board members convene to address splits and ties and to determine officer record order of merit, numbers identifying where members lie on a list of potential candidates from 1 to end. Additional consideration takes place a final time to review packages falling at the brink of the approved candidates, and a final list of commander candidates is published. After the board, hiring officials may request one-on-one interviews with the commander candidates before matching candidates to specific positions, although this is not required.
Within the Air Force, a plethora of doctrine, regulations, instructions, manuals, and technical orders exist, ranging from how to properly use a chair to developing a strategy for modern international warfare with near-peer threats. Still, one of the most critical aspects of the military foundation is neglected — leadership in command. Currently, no Air Force doctrine discusses the importance, necessity, and required qualities of future commanders. As of 2022, the Air Force focused on Airmen qualities as a whole and relied on past job performance as an indicator of future leadership success. It did not identify the leadership qualities or attributes expected from commanders. As such, specific leadership qualities are not evaluated when considering officers for promotion and command positions.
The lack of specified qualities is a new phenomenon in the Air Force. From 1947 to 2004, the Air Force published regulations on leadership focusing on specific leadership attributes the force required, which was based on Air Force wars and missions. But, it last identified leadership attributes in 2015. By contrast, the U.S. Army doctrine page demonstrates leadership emphasis by having a publication for leadership and a separate publication for mission command. The Army clearly defines leadership traits and competencies through doctrine, identifying the leadership requirements model as the basic component of what a leader is and what a leader does.
The lack of clearly defined command expectations means there is an inability to quantify essential data related to leadership qualities, thus enabling unstandardized officer performance assessments based on rater personal preferences. This bleeds over into the promotion and commander selection board process. Moreover, the current system does not include all possible factors that should be explicitly assessed during screening boards, such as results of investigations in which members are accused of wrongdoing, inspector general, equal opportunity allegations, and permanent change of station or assignment earlier than expected for negative reasons that did not rise to official removal of command. Not including this data enables officers with bad performance to continue competing for command positions, and their negative behavior eventually becomes obsolete when board members no longer remember negative instances not formally documented in performance reports.
Enduring Subjectivity
This subjectivity of officer selection is a known element of the process, where the opinions and knowledge of a particular officer in question may or may not be discussed in boards to make final determinations of selections. Consider the case of Air Force Major General Dawn Dunlop, who continued to advance her career, despite having been fired for creating a toxic work environment. In a possible scenario, Lieutenant Colonel Dunlop’s previous toxic actions did not rise to the level of formal action, and Board members either would not have had knowledge on Dunlop or opted not to provide the information to the board due to allegiance or favoritism.
Compounding this problem is the fact that the board selection process is exhaustingly long and tedious. The new officer performance brief format decreases the time it takes to write the report, but it does not reduce the time it takes for screening boards to review the packages or improve clarity on qualities the members possess. Board members with limited time cannot comb through records and instead typically rely upon two lines per performance report, about five to ten lines out of 50 total lines for a Major (O-4) package, to decide on the worthiness of selection. This process leads board members to pick commander candidates by reviewing limited data and making judgment calls based on their previous experiences, personal beliefs, values, and intuition, leading to the selection of officers who are “good enough” but not necessarily the best. Additionally, board members have to rely on performance reports that, until recently, failed to account for the skewed, subjective, and sometimes deceptive officer stratification process that often relied upon timing, officer visibility, the “good old boy system,” and officers maintaining a “halo effect.”
In total, this process determines command selection based on cognitive biases, or more specifically, a process that adheres to social identity theory. Board members make unconscious errors and inadvertently show partiality towards individuals based on the members’ impression of an individual or if the board member considers that an individual is “in” a certain network of personnel. Because they have limited time and can only use the documents in an officer’s package, board members use shortcuts, also known as heuristics, based on beliefs, past experiences, and intuitive mechanisms, to help them select individuals.
This bias is understandable: a board member does not have the luxury to analyze stratifications, push statements, and decorations deeply and must rely on his or her training and understanding of the evaluation system to make a quick decision based on intuitive mechanisms related to his or her experiences. He or she is engaging in a process to reduce uncertainty in an attempt to build a cohesive group. But, ultimately, it is a groupthink process built on subjectivity, not objectivity.
The result is a system that does not assess candidates to select the best officer for the job; rather, it selects officers who are “good enough.” The impact is evident: in 2023, a cursory Google search identified at least 13 Air Force news stories about officers relieved of command due to loss of confidence, shortfalls in personal conduct, inability to complete their assigned duties, criminal misconduct, etc.
The Army, in comparison, has put in place a system “designed to ensure fairness and meritocracy” that does not allow these officers to take command in the first place. With its Battalion Command Assessment Program, the Army identified 34 percent of its officers not ready for command, who would have been chosen for command in its legacy system. More importantly, Army officers have embraced the new system with 94 percent saying it is better than the prior system and 97 percent of officers wanting it to continue. A secondary benefit is its safeguard against general officers attempting to bypass the system and promote officers unfit for command.
A Better Alternative
The officer development, selection, and assessment processes should be revamped if the Air Force is serious about promoting the best candidates for command positions. The Air Force needs a process wherein it can consider all known, relevant, and consistent data points in its assessment to make the best decision with maximum results and project future success.
This process involves using an expected utility decision-making model, in which board members and the Air Force assess officers with a more quantitative view of their performance and future potential. Expected utility falls under the rational choice approach, in which individuals make comprehensive, compensatory, and predictable decisions. This approach is methodical, and decision-makers aim to maximize gains and increase the probability of success by comparing the expected utility values of all prospective decision factors. Notably, military officers are a unique category, distinct from civilian leaders, but generalizable observations have been made for them.
Using expected utility to identify favorable elements in leadership and officer success, the Air Force could improve, streamline, and increase consistency when selecting the best officers for command positions. To improve the selection process, the Air Force first needs to develop clear guidelines for what it looks for in commanders and incorporate those critical elements into Air Force doctrine. The factors identified by senior Air Force leaders must be readily accessible and at the forefront of discussions with all officers and future commanders.
Additionally, the Air Force needs to develop commander assessment programs similar to the Army Command Assessment Program and Navy command screen process before a commander is matched with a command billet and fully re-develop the assessment process to include leadership milestones. The Command Assessment Program, in particular, has had a history of success and adaptability. This would mean assessment program variables that include all the current officer performance brief data, the addition of established leadership qualities to evaluate (such as “understands and executes appropriate risk tolerance,” “adaptability,” and “decision-making in crisis”), command assessment results, cognitive, non-cognitive, physical, written and verbal assessments, psychological interview results, demographic information, and previous key billets filled.
The combination of Air Force doctrine-driven variables and the variables derived from a developed commander assessment program could then be used to produce a model, forecasting what factors are evident in a successful commander. Models are flawed, and so it would need to be adjusted, as necessary, based on a constantly increasing pool of results. With this model in place, the most deserving individuals could then be identified for command consideration. A model would also assist in avoiding discrimination and ensuring equality of promotion opportunity. The evaluation process could then be made fast and fair.
In an age where AI is everywhere, clearly defined and collected variables will allow for a faster, more efficient, and more streamlined selection of officers to lead. AI is not infallible, but it can be used as a tool to reduce subjective interpretation of data, improve decision-making, and increase fairness in decision-making. As the Army does with its Command Assessment Program and in building its army leadership requirements model, “bad” data would be removed through a time-series process of identifying the variables correlated with both successful and unsuccessful commanders (for example avoiding selection bias and survivorship bias). Machine learning algorithms learn only to consider relevant data for their decisions, which improve their predictive accuracy in line with rational choice theory. Using AI would allow for the removal of cognitive biases when assessing packages, remove individual preferences, and, in some instances, help promote qualified personnel from disadvantaged populations. AI could be a tool, subject to human inputs and without removing humans from the final decision, to give board members more time, data, and analysis to make better, more informed decisions.
Change is uncomfortable but necessary to ensure the Air Force is ready for future conflict. Today’s Air Force command assessment process is flawed, but it can be fixed. If the Air Force adopts a more complex but objective expected-utility decision strategy to fully assess commanders before they assume command, the number of unprepared, unsuitable, and inappropriate leaders would decrease while the number of competent, authentic, and diverse leaders would increase.
Major Maria Patterson has 12 years of dedicated Air Force service and has served as a three-time graduated commander in both garrison and deployed missions for the Air Force Office of Special Investigations. Major Patterson has a robust background in leadership development, has served as a Field Operating Agency officer mentor, has been present at officer promotion boards, and received board out-briefs during her career.
Dr. Bradley Podliska is an Assistant Professor of National Security Studies at Air University. He has a Ph.D. in Political Science from Texas A&M University and is the author of several publications, including Fire Alarm: The Investigation of the U.S. House Select Committee on Benghazi (Lexington Books, 2023). Follow Bradley on X/Twitter: @BradleyPodliska.
The views expressed are those of the authors and do not reflect the official views of the United States Air Force, nor the Department of Defense.
Image: Airman 1st Class Alysa Calvarese
Comments are closed.