Path dependency in policymaking: why does it arise?
On how the past constrains the present in policymaking
Defining path dependency
Policymakers wield the power to develop and implement strategies that drive economic growth and innovation. While many politicians hold significant influence over policy decisions, we’re becoming increasingly aware of the constraints – the ‘rules of the game’ – that policymakers face in their policy choices. These include their incentives and resource constraints, but also their duties to voters (political business cycles), systemic checks and balances, and the enigmatic forces of bureaucracy.
History also binds policymakers through path dependency – constraints on the set of present policy choices due to past decisions. Policy can play a major role in helping or hindering innovation, so the stakes are high. Understanding the effect of path dependency on policy and strategy is therefore crucial.
Path dependencies exist everywhere in life:
The standard English keyboard (QWERTY) is a perfect example of path dependency. Designed in 1868 to prevent typewriter jams, we’ve grown accustomed to this particular design. Even though some challengers to QWERTY (e.g. Dvorak) may be faster, more convenient, and easier to learn, the equilibrium has settled on QWERTY. We’ve locked ourselves in.
In language, the phrases ‘left-wing’ and ‘right-wing’ in modern political discourse originate from the French Revolution, in which the two factions, revolutionaries and aristocratic supporters, sat respectively on the left and right of the French National Assembly. Would our politics be less divisive if we adopted a different political terminology that moves beyond a linear view of the political spectrum? We wouldn’t know as ‘left-wing’ and ‘right-wing’ have been so entrenched in our everyday language!
Knowledge also appears to exhibit path dependency. In writing about our knowledge of economics, Mark Blaug, an economist, writes that ‘what we now know about the economic system is not something we have just discovered, but it is the sum of all discoveries, insights and false starts in the past’. More widely, the enterprise of knowledge-building is based on paradigms, which are broadly accepted collections of assumptions for the practice of science. The discovery of facts and knowledge about the world by scientists and researchers is path-dependent on that corpus of assumptions and established practices.
Similar path dependency exists in growth and innovation policy. For instance, policies to renovate transport networks such as the London Underground, built in 1863, are highly path-dependent, as the immense fixed cost of demolishing and rebuilding a transport network greatly increases the cost of making changes after initial infrastructural development. Incentivisation policies can have long-lasting effects: A survey of small farms in Peru and Brazil found that past successful policy incentives (e.g. credit incentives for mechanisation) that were effectively implemented can have a long-lasting influence on present land use plans.
Why does path dependency arise in policymaking?
Policymakers often face path dependency in policymaking for several reasons:
Existing policies self-perpetuate.
Policies may be self-perpetuating. Paul Pierson, a professor of political science, attributes such self-perpetuation to inertia, in which ‘positive (read: additional) feedback may lead to a single equilibrium’. For instance, governments establish systems around government policies and strategies for effectively maintaining and implementing a particular decision. Departments prepare application forms and paraphernalia; governments set up offices; leaders hire individuals to manage new programmes. For many institutions, path dependency manifests as a ‘vendor lock-in’ problem, in which businesses or governments grow dependent on particular vendors, gradually accumulating switching costs to a point at which switching vendors becomes too costly to be justifiable. For example, the UK government struggles with path dependency for procuring cloud contracts with leading cloud service providers such as AWS.
More importantly, citizens’ expectations, decisions, and customs are affected by past policies, which constrain policy choices in the present. Researchers Isti Hidayat, Claudia Yamu, and Wendy Tan document such path dependency for Jakarta’s transportation policy. In the 1980s, the local city government built to connect housing enclaves across Jakarta. Nowadays, even though the local government has thoroughly invested in pedestrian improvement projects and mass commuter rail to reduce private road usage, the ‘organisational routine’ of the convenience of private is too stagnant; policy changes are correspondingly ineffective. In all cases, we see that a policy choice in the past can alter the calculus for what constitutes an effective policy today. Self-perpetuation is at play.
The policymaker lacks knowledge, fears uncertainty, or sees no incentive for change.
Policymakers may lack the know-how to consider and implement policy suggestions, leading to path dependency. Past policies are easier to implement because systems and structures are likely in place to maintain such policies. There may also be more data about the efficacy of the policy – in other words, more certainty. In contrast, new policies are uncertain; ambiguity-averse institutions and policymakers, who have a stronger preference for known rather than unknown risks, may choose to avoid these new policies that may be more effective but instead stick to what they know. The past consequently restricts the present because policymakers are incentivised to adhere to familiar and established strategies. Sticking to past policies is the natural course of action within political institutions.
Gaining the buy-in for sudden policy changes can be difficult.
Changing a policy or altering a decision is difficult after it has been enacted. For instance, a government may commit to a system of unemployment benefits during a recession, which is an easy choice at the time. But removing these benefits when the economy returns to ‘normal’ may be much more politically and ideologically difficult. For instance, a ‘welfare state’ government may act with an eye to the ballot box but also encounter protests while reducing subsidies or benefits as removing welfare is publicly perceived as less acceptable.
Similarly, in urban planning or development, NIMBYism (‘not in my backyard’, encapsulating individuals that object to infrastructural development or reform in their ‘backyard’) may also pose an obstacle to future change. Researchers have documented NIMBYism as a major barrier to change at public hearings for planning in San Francisco or to local economic growth policy in the UK. Past policy choices, especially those that directly mix with individual livelihoods and stability, develop new expectations from individuals, fostering status quo bias and constraining policies open to policymakers.
Should we be concerned?
Path dependencies are everywhere in policymaking. But path dependency in the context of growth and innovation policy, in which significant fixed-cost infrastructure or innovation incentive schemes can have long-reaching consequences and constrain the feasible policy set in the future, remains a relatively neglected issue in the political science and innovation literature. In a future essay, we’ll explore whether path dependency in policymaking is something we should be concerned about, as well as solutions for reducing negative path dependencies.
