Few of us may have realized it, but we celebrated an important policy birthday this past Sunday: the fourth anniversary of the passage of the American Recovery and Reinvestment Act, popularly known as the Obama administration's economic stimulus program.

My colleagues and I at George Mason University have issued a report taking stock of the management of this complex initiative. This reflection revealed a grave disconnect between economic reality and political perception. Looking at the record, the over $800 billion in federal spending and tax cuts has in fact borne fruit. The Congressional Budget Office reports that the stimulus created or saved as many as 4.7 million jobs. And as bad as state-government deficits have been, they would have been far worse had the federal stimulus not provided billions in grants-in-aid to shore up state treasuries.

Furthermore, unlike with earlier stimulus programs in prior recessions, Congress and the Obama administration acted with remarkable speed. The 2009 stimulus became law just 14 months after the recession officially began in December 2007--significantly faster than the average of 27 months between the beginning of past recessions and the enactment of prior stimulus initiatives.

Yet polls show that most of the public feels the stimulus did not work. Ironically, this outcome has happened irrespective of unparalleled efforts by this administration to make the spending and results of the stimulus programs transparent to the public. Far from promoting public confidence, however, the unprecedented information about job creation and spending rates has enabled a highly mobilized set of interests in the blogosphere to promote those results that suit their biases.

A comparison to the New Deal is helpful to explain the disconnect between the economics of the stimulus and its politics. The New Deal job-creation programs--the Works Progress Administration, the Public Works Administration and the Civilian Conservation Corps, among others--enjoyed tremendous popular support, as did their primary advocate, President Franklin Roosevelt. Each of these programs developed a strong political following among grateful clients and prospective clients who identified these as a safe harbor in a horrible economic storm.

What did the New Deal and President Roosevelt have that we lack today? First, the sheer depth of crisis and desperation motivating the New Deal made many institutional innovations politically possible that before then were unthinkable, generating bipartisan support in the first few years of FDR's term.

Second, FDR had enormous personal popularity that translated into support from an overwhelmingly Democratic Congress. Popular presidents were able to do this at a time when much of the electorate voted a straight party ticket, influenced by the leading man at the head of the list. By contrast, Obama and most postwar presidents have had to cobble together majorities in a more perplexing and polarized party system.

Finally, unlike today, the important jobs programs of the New Deal were direct centralized federal programs in which the allocation of resources and benefits were largely made by federal officials, often political appointees of the president. The millions who obtained relief and work from such programs as the Civil Works Administration and WPA, as well as many others seeking relief, learned to train their hopes on the White House for deliverance.

In some respects, centralization was borne of necessity. The New Deal was forced to create new programs and administrative institutions from whole cloth, hurriedly building an administrative establishment and capacity to match the breadth of the new roles taken on by the federal government. Unlike today, most states and localities at the time had little administrative capacity or political will to deliver these expanded responsibilities.

By contrast, the current stimulus is delivered using a highly decentralized structure characteristic of nearly every major federal domestic initiative of the postwar era. Reflecting widespread disaffection with centralized federal governance, the expansion of national policy over the past 65 years has been carried not by legions of federal bureaucrats but rather by a wide range of third parties, most prominently state and local officials.

During the first quarter of the stimulus program, there were more than 130,000 state and local governments, nonprofits and private firms that reported receiving stimulus funds under federal grants, contracts and loans--a figure that excludes thousands of others receiving subsidies under tax programs and entitlements. In most cases, those federal funds traveled down a complex labyrinth, often passing through several layers before reaching the ultimate point of service delivery. To make matters worse, the primary impact of the stimulus in many communities was to prevent layoffs rather than hiring new people; the prevention of bad outcomes is not easily appreciated.

As a result, there is no clear line of sight for the public to attribute credit to the president, when governors, mayors and other political actors are likely to be competing for public approval. In fact, the public is often confused about whom to credit or blame when so many hands are stirring the pot.

The problem with the stimulus is not, as many critics have claimed, that President Obama is centralizing too much power at the national level. Rather, it is that he has inspired broad expectations while being consigned to use a highly decentralized governance system which disperses credit to others and centralizes blame in the presidency itself.

The political dilemma facing the Obama administration reflects a clash between the public's desire for stronger national programs and our continued attachment to our long traditions of limited, localized governance. It takes an exceptional leader in exceptional times to resolve the fundamental paradox of the American view of government: "Get the government off our backs" and "There ought to be a law."