Kerr and Jermier (1978) used the term substitutes for leadership to describe the characteristics that make leader behaviors irrelevant, unnecessary, and redundant. For example, some characteristics that help subordinates better understand their roles and work processes, or allow them to obtain feedback and rewards from sources other than their managers, function as substitutes for leadership (Kerr & Jermier).
In my view, we can associate many phenomena to leadership but it is non-sense to discuss about a substitute for leadership. Substitutes in strategic management are those products and services which satisfies the same needs. The specific needs satisfied by leadership are vision, motivating people, change dynamics and a holistic understanding of the future. Any manager able to satisfies these specific needs may become a good leader but not a substitute for a good leader. Thus, in my view there could be no logical relationship between leadership and substitutes for leadership. However, we may consider a logical relationship between the leadership and organizational ideology since leadership is always supported by a certain organizational philosophy in terms of vision and values.
From what I understand, there are many behaviors associated with substitutes for leadership. I was curious if you had a specific behavior you were considering?
For example, when you look at the subscales from Kerr and Jermier there are two areas that are intrinsically motivated by the employee. In this case, there is no need for Leadership Empowerment Behavior. Which would indicate that there is not logical relationship between the three variables.
However, if you look at the subscales that are related to organizational ideology (training, staff functions, work group cohesiveness, and being allowed to work independently). Then, a case could be made for a logical relationship. In this case, there is at least a case for further exploration to determine if Leadership Empowerment Behavior is a part of the relationship.
Fundamentally, the answer is "maybe". There are a lot of latent variables in this scenario. The results would depend on how the variables moderate or mediate each other and whether a researcher wants to make a case to explore them.
The main point to verify the logic of the proposed relationship is to define the theoretical framework. The purpose of a theoretical framework is to serve as a guide for predicting how your variables should interact, and without a strong framework your hypotheses and results are open to many alternative explanations.