A comprehensive study conducted by Martin Gilens and Benjamin I. Page.

IN SUMMERY:

A study revealing that economic elites and organised groups representing business interests exert substantial independent impacts on U.S. government policy.

Conversely, average citizens and mass-based interest groups demonstrate little or no independent influence on policy outcomes.

The majority does not rule in the causal sense,

providing strong support for ELITE THEORY.

THE THEORY

The study examines four principal theoretical traditions concerning policy influence:

Majoritarian Electoral Democracy: 

This theory posits that government policies primarily reflect the collective will of average citizens, empowered by democratic elections.

Vote-seeking parties or candidates in a two-party system are expected to converge at the midpoint of citizens' policy preferences, leading to public policy that aligns with the median voter. Such theories predict a positive, significant, and substantial influence of average citizens on policy, with other actors having little or no influence.

Economic-Elite Domination: 

This tradition argues that U.S. policy-making is predominantly shaped by individuals possessing substantial economic resources, such as high income or wealth.

These theories anticipate a positive, significant, and substantial influence of economic elites on policy. While some independent influence by average citizens on non-economic social issues, or by business interest groups, may be acknowledged, the emphasis remains on wealthy individuals.

Majoritarian Pluralism: 

Rooted in concepts like James Madison's factions, this theory suggests that struggles among diverse organised interest groups and business firms lead to policies that are largely representative of the citizenry's needs and interests as a whole.

It implies that the average citizen's wants are reasonably well served by the outcomes of interest-group struggle. This perspective predicts that organised interest groups, collectively, would have positive, significant, and substantial effects on public policy, often reflecting average citizens' preferences.

Biased Pluralism: 

An important variant of the pluralist tradition, this theory posits struggles among an unrepresentative universe of interest groups, where corporations, business associations, and professional groups predominate.

It predicts that both the direction of interest-group conflict and the resulting public policies tend to favour the wishes of business-oriented entities. In this view, organised interest groups have far more influence than average citizens or individual economic elites, with business-oriented groups playing the major role.

SPOILER > THIS IS THE WAY

METHODOLOGY 

The research employed a unique dataset compiled over many years, including measures of key variables for 1,779 policy issues proposed between 1981 and 2002. The policy issues included were those on which a national survey of the general public had asked a favour/oppose question about a proposed policy change and met specific criteria, tending to concern matters of relatively high salience.

Average Citizen Preferences: 

Measured by policy preferences at the fiftieth income percentile (median-income survey respondent), serving as a proxy for the median non-institutionalised adult American.

Economic Elite Preferences: 

Proxied by the policy preferences of Americans at the ninetieth income percentile, which, while not representing the "truly wealthy," were found to capture useful information about their views. It is posited that this proxy likely leads to an underestimation of the impact of economic elites.

Organised Interest Group Stands: 

Measured using a "Net Interest Group Alignment" index. This index involved coding engaged interest groups as "strongly favourable," "somewhat favourable," "somewhat unfavourable," or "strongly unfavourable" to a proposed policy change.

It included groups from "Power 25" lists and ten key industries with the highest lobbying expenditures, weighting positions and taking logarithms to account for diminishing returns. Separate indices were also computed for mass-based and business-oriented groups.

Dependent Variable: 

Whether the proposed policy change was actually adopted within four years after the survey question was asked.

FINDINGS 

Average Citizens: 

The estimated impact of average citizens' preferences dropped precipitously to a non-significant, near-zero level.

Ordinary citizens have little or no independent influence on policy at all.

Even overwhelming majorities favouring a policy change obtained that change only about 43 per cent of the time.

Economic Elites: 

Economic elites were estimated to have a quite substantial, highly significant, independent impact on policy.

They stand out as quite influential—more so than any other set of actors studied—in U.S. public policy-making.

Organised Interest Groups: 

Organised interest groups were also found to have a substantial independent influence on policy.

These multivariate results strongly indicate that economic elites and organised groups representing business interests have substantial independent impacts on U.S. government policy.

Average citizens and mass-based interest groups have little or no independent influence.

This provides substantial support for theories of Economic-Elite Domination and for theories of Biased Pluralism, but not for theories of Majoritarian Electoral Democracy or Majoritarian Pluralism. See ELITE THEORY.

CORRELATIONS

A key finding is that the preferences of average citizens are positively and fairly highly correlated with the preferences of economic elites across issues, meaning they often desire the same outcomes from government. This explains why ordinary citizens might frequently appear to "win" even if they have no independent effect on policy-making, as elites (with whom they often agree) are the ones who actually prevail.

However, net interest-group stands are not substantially correlated with the preferences of average citizens.

When distinguishing between mass-based and business-oriented interest groups:

• Both mass-based and business-oriented interest groups show positive and statistically significant influence coefficients.
• Crucially, the coefficient for business groups is nearly twice as large as that for mass groups.

The evidence strongly indicates that theories of Biased Pluralism are more descriptive of political reality than Majoritarian Pluralism. Existing interest groups do not effectively serve as transmission belts for the wishes of the populace, and "potential groups" do not fill this gap.

IMPLICATIONS 

The findings present troubling news for advocates of "populistic" democracy, who desire governments to respond primarily to citizens' policy preferences. In the United States, the majority does not rule in the causal sense of actually determining policy outcomes. When a majority of citizens disagrees with economic elites or organised interests, they generally lose. Furthermore, due to the substantial status quo bias built into the U.S. political system, even large majorities favouring policy change often do not achieve it.

While Americans enjoy features central to democratic governance, such as regular elections, freedom of speech and association, and a widespread franchise, if policy-making is dominated by powerful business organisations and a small number of affluent Americans, then America's claims to being a democratic society are seriously threatened.

While elites may argue they possess greater expertise or seek the common good, it is the ordinary citizens who generally understand their own values and interests, and their expressed preferences warrant respect.

It cannot be assumed that informational expertise in elites is coupled with an inclination to transcend self-interest or work for the common good. The public is likely a more certain guardian of its own interests.

The study's limitations, such as the imperfect proxies for truly wealthy Americans and the limited scope of included interest groups, suggest that the actual influence of economic elites and organised groups may be even greater than estimated.