A summary of key concerns and some solutions
No alternative to LFA The Logical Framework Approach (LFA) has been around for well over 40 years. It has been enormously successful in terms of widespread use, at least in part due to its simplicity. It has been much less successful in international development in terms of sustainable results and appreciation by project staff. Yet, it is an honest attempt to address some of the planning and implementation issues that beset the sector. Probably some of the critique leveled against it, raises concerns that existed will before its introduction. If we admit that there is no alternative, the question remains how serious the critique is and how it may be addressed.
Basic design LFA is based on a mechanistic, linear management model that presupposes perfect knowledge of cause and effect and total control over implementation. Once the links between causes and effects are known with sufficient certainty it is possible to formulate and implement a strategy that is very likely to be successful. The trouble starts with complex, changing situations involving multiple organizations or people: the links between causes and effects become highly uncertain as do the sequences in which they occur.
LFA in complex situations …. tends to force a consensus on causal logic based on superficial assumptions in order to fit the simple matrix of the LogFrame. Whatever logic there was at the start is lost due to forcefully fitting it into the LogFrame matrix (“jamming”), leading to a so-called “lack-frame” that does not do justice to the complexity of the real situation at hand. As a result LFA becomes a tool of rigid planning, where flexibility was required. The result is a so-called “lock-frame.”
Bureaucratization Linear management is the bread and butter of bureaucrats and hierarchic structures generally. So LFA contributes to the bureaucratization and top-down control of aid. The tunnel vision imposed by LFA confirms often manifest managerial belief of self-importance and strengthens the audit focus of LFA. This drives a vicious cycle of ever increasing demand for accountability and mistrust. For many donors, LFA is a requirement for any project to reach approval status. This results in preconceived projects being squeezed into the LogFrame, ignoring logical levels between levels or any prior assumptions and turning it into a “logic-less Frame”.
Top-down control In LFA, much use is made of expert planning at the start of inter-organizational or multi-stakeholder projects or programmes. The experts are controlled by terms of references (ToR’s) prepared by donor decision-makers. Further control is exercised by the hiring regime. In addition, the LogFrame itself can be quite restrictive because its focus on expected results favours a quantitative approach in terms of products rather than a qualitative focus on process and values. Unless the expert is encouraged by the ToR to set up an extensive dialogue he will be constrained to use his experience to actually replace this dialogue. For a concept map of the drawbacks of LFA click here.
Project cycle management (PCM)… was invented to address the sustainability concerns about LFA. PCM is an attempt to reduce the product focus of LFA, to improve project preparation, and to encourage a better decision-making discipline. If properly applied, it is probably a great improvement over previous LFA practice, but it suffers of two defects: a continued domination by LFA and a continued problem orientation, thus effectively preventing local learning.
Outcome Mapping (OM) … was not designed with LFA in mind, but could correct some of its faults, because it has a strong process focus. It considers that real development demands a change in behavior of key development partners. A project managed with OM has activities to support so-called boundary partners to produce the desired type of behavioral change. Boundary partners are called thus, because they represent the boundary of what the project can control. The “actual” development is considered to be beyond the control of the project. OM takes in learning, complexity, partner perspectives, and non-linearity. Various models for integrating LFA with OM or vice versa have been proposed.
Other alternatives In their note on programming for developmental complexity, Hummelbrunner and Jones (2013) examine ways for admitting systems thinking into planning and strategy development. They too mention Outcome Mapping as one of the approaches that could be used. But they also list a host of other approaches, including scenario technique, conditional planning, milestone planning, assumption-based planning, boundary planning, and adaptive strategy development. Some of these may well be usefully combined with LFA. Hummelbrunner (2013) further mentions social network analysis as a way for dealing with uncertainty in LFA more effectively.
Critique The question remains to what extent these alternatives are able to address the basic concerns of the human problem situations or “wicked problems” that underlie the impact challenges of international development. How many tools and approaches do we need (and therefore do we need to acquaint ourselves with) before we can start working systemically? According to Rittel and Webber (1973) social system problems involve systems of systems that are interacting and lack a problem centre. As a result goal finding and problem definition will always remain elusive.
Ideal planning system The idealized (and impossible!) planning system described by Rittel and Webber to address this challenge involves an ongoing simultaneous process of considering the systems’ multiple values (or world views), searching goals, identifying problems, inventing strategies, simulating alternatives, forecasting contextual changes, and monitoring changes in relevant conditions. This points to a situation where planning and inquiry cannot be separated. Systemic planning is an approximation of the idealized planning system just described.
Systems approach A well-thought-out generic approach that does both systemic planning and systemic inquiry was developed by Churchman (1968). He termed it “the systems approach” before anybody else did. Perhaps this approach was insufficiently practical for business purposes to facilitate its breakthrough in the development sector. But it does seem to have the necessary ingredients for sweeping in many of the systemic considerations identified by Rittel and Webber and that are now lacking in LFA, with or without accompanying process-based or strategy-oriented methods.
Tentative conclusion LFA on its own is good for basic planning, budgeting, and accountability in simple situations, but it is very weak systemically speaking, which becomes important in more complex situations. PCM and OM are able to correct this to a large extent. For strategy work, there is a choice of dedicated methods. For catering to specific systems aspects the systems approach (SA), or its critical derivative CSH, seems the best option. It is not commonly known that SA also has a few “hard-nosed” elements, e.g. where the input-output analysis of project components is concerned. SA & CSH are at their most powerful in dialectical valuation, in which it combines the fundamental systems concepts of perspectives and boundaries.
- Hummelbrunner, R. (2010). Beyond Logframe: critique, variations and alternatives. In Beyond logframe: using systems concepts in evaluation. http://users.actrix.co.nz/bobwill/Beyond%20Logframe.pdf#page=8
- Hummelbrunner, R., & Jones, H. (2013). A guide for planning and strategy development in the face of complexity. http://www.odi.org.uk/sites/odi.org.uk/files/odi-assets/publications-opinion-files/8287.pdf
- Rittel, H. W. J., & Webber, M. M. (1973, June 1). Dilemmas in a general theory of planning. Policy Sciences, 4(2), 155–169. http://www.uctc.net/mwebber/Rittel+Webber+Dilemmas+General_Theory_of_Planning.pdf