As the novel coronavirus has spread across U.S. cities and states, so have public policies aimed at stopping the pandemic. Sometimes, states learn from others’ policy successes. Other times, they simply copy neighbors or even compete against them.

So far, state and local government policy choices to stem the tide of covid-19, the disease caused by the virus, look more like simple imitation and harmful competition than thoughtful learning and innovation. But we see opportunities for learning going forward.

Here’s why.

How state governments initially responded

In mid-March, governors in all 50 states acted within days of one another to declare states of emergency, a move that enhanced their powers to combat the coronavirus. Starting with California, states then began to issue stay-at-home orders to reduce opportunities for their citizens to transmit the virus. Soon, 45 states had fully or partially adopted stay-at-home policies, except Arkansas, Iowa, Nebraska, and North and South Dakota. Compared to the rates at which states ordinarily adopt other important policies, stay-at-home orders spread strikingly fast.

But such similarities across states soon break down; beyond them, we see a patchwork of state and local government responses. Weeks after the World Health Organization declared covid-19 a pandemic, states still faced shortages of the equipment necessary to fight the virus. Lacking sufficient ventilators, respirators and face masks, states began competing with one another to obtain the equipment in the open marketplace. States have also gone their own ways to obtain testing kits and to decide whom to test.

Here’s how policies typically spread in the U.S. federal system

These responses reveal both the benefits and costs of the U.S. federal system, which couples a national government with a broad array of state and local governments. When a federal system works well, governors and mayors tailor responses to local needs, experimenting and learning about other cities’ and states’ best policy choices. However, federalism can result in haphazard policy responses and harmful internal competition for resources.

Our research identifies three primary mechanisms for how policies spread, or what social scientists call “policy diffusion”: imitation, competition and learning. Imitation involves state or local officials identifying what another government has done and quickly adopting that policy, without giving much consideration to how well it fits at home. Competition occurs when governments act to remain one step ahead of or catch up with others. In contrast, learning involves governments observing what others are doing, then determining which of those policies would work best for their own states or localities.

We see evidence of all three mechanisms in state and local responses to the pandemic.

The limits of imitation and competition

When governments imitate one another, policies can spread quickly, with states and cities just copying what others have done. Despite the advantage of speed, imitation does not allow policymakers to determine whether such policy choices are appropriate for their own particular circumstances. Much of the policy response to the covid-19 outbreak was this fast imitation type by necessity. States quickly imitated one another in policies such as school closings and stay-at-home orders. But if such an approach were to continue, imitation could lead to the economy reopening too quickly. Once some states increase their capacity for testing and contact tracing, their changing policies might place pressure on others, even those that are unprepared, to act similarly.

Competition can have the benefit of encouraging experimentation. Acting as laboratories of democracy, to use Supreme Court Justice Louis Brandeis’s famous phrase, states have explored different ways to extend unemployment and Medicaid benefits, or to support small businesses. Such actions might offer their economies advantages over neighboring states, while also raising policies that other states may wish to adopt. At the same time, states have been competing — potentially destructively — for essential medical supplies.

There’s still a chance for states to learn

Learning, of course, takes time, and this pandemic isn’t offering much of that. But research suggests that states and localities actually can learn fairly quickly. We see three paths forward.

First, when governments pool their resources or coordinate their responses, they are more likely to learn about best policy options. Organizations like the National Conference of State Legislatures or the National Association of County and City Health Officials can facilitate such coordination, much as the National Governors Association has been trying to coordinate the purchase of protective equipment across states. States are also beginning to organize regional coalitions on their own.

Second, cities and states can learn from the federal government’s actions and statements. If Washington were providing clear, consistent and coherent signals, subnational officials could experiment and learn more quickly.

Third, state and local governments with greater expertise and capacity are more likely to learn about which actions have been successful elsewhere and then usefully adapt them to their own circumstances. States with expertise can marshal this knowledge; other states can assemble and rely upon experts to study the effects of earlier policy experiments. Already, evidence about the success or failure of different policy approaches is starting to emerge from locations like New York, which saw the most dramatic increase in coronavirus cases, and Florida, which was slower to impose restrictions.

Even while the pandemic is moving quickly, different cities and states are being hit at different times. That offers more opportunities for officials to learn from their peers’ decisions about health policies, about limiting damage to the economy, and eventually about how to emerge from the worst of the pandemic.

Charles R. Shipan is the J. Ira and Nicki Harris Professor of Social Science at the University of Michigan, where he is a faculty member in the department of political science and the Ford School of Public Policy.

Craig Volden is a professor of public policy and politics at the Frank Batten School of Leadership and Public Policy at the University of Virginia, and he is a co-director of the Center for Effective Lawmaking.