Media Markets, Special Interests, and Voters Leopoldo Fergusson U Andes October 27, 2011 Fergusson (U Andes) Media & Special Interests October 27, 2011 1 / 24
Introduction Major concern in democracies Influence of special interest groups Some US Examples The “Keating Five” Scandal (1989): Five Senators accused of intervening to delay regulation in 1987, acting on behalf of Charles H. Keating, head of Lincoln Savings and Loan Association, who had contributed to the Senators’ political campaigns. The recent financial meltdown: Political power of Wall Street in Washington as a fundamental cause: “Campaign contributions and the revolving door between the private sector and government service gave Wall Street banks influence in Washington” (Johnson and Kwak, 2010). Recent “mid-term elections:” Spencer Bachus (R.-Ala), received over a million dollars from ”banks, insurance companies and auditors (...) “And wasting no time (as chairman of the House Financial Services Committee) Bachus sent a letter to the Financial Stability Oversight Council that reads as if dictated by bank lobbyists” Fergusson (U Andes) Media & Special Interests October 27, 2011 2 / 24
Introduction How can the influence of interest groups be tamed? The role of media In a democracy, the most basic safeguard are elections. If influence on a particular candidate is (or appears) improper → vote to punish. But for this, voters need information, and mass media is the main source. For this reason (among others) the presence of a free, independent mass media is a key component of democratic political institutions. Many believe mass media indeed played this role historically Example: Early XXth Century US, from the ”Gilded Age” to the ”Progressive Era”. Free press and journalists (”muckrakers”) exposed the excesses of Robber Barons as well as corruption in local and federal politics. Fergusson (U Andes) Media & Special Interests October 27, 2011 3 / 24
Introduction The questions Mass media countering special interests 1 Can mass media counter the influence of special interest groups? 2 In particular, can it help voters respond to potentially improper capture of candidates by special interests as signaled by campaign contribution patterns? If so, how important is this e ff ect? Fergusson (U Andes) Media & Special Interests October 27, 2011 4 / 24
Introduction The approach Compare voters’ response to campaign money across di ff erent media markets County-level support for US Senate candidates, 1980 to 2002. Variation in media exposure: TV markets’ structure [map]. Media market In-state Out-of-state Coverage of state politics: High Low Variation in ”capture” of politicians by special interests: campaign contributions data. Campaign finance pattern: Concentrated Dispersed Politician ”capture”: Likely Unlikely Fergusson (U Andes) Media & Special Interests October 27, 2011 5 / 24
Introduction Pattern of campaign contributions is a good proxy for capture by interest groups Some anecdotal evidence Individuals often do not know the pattern of campaign contributions. But even if this is the case, a candidate with a more concentrated pattern of contributions is more likely to be pinpointed by journalists or his opponent as potentially “captured.” Such candidate is likely to figure in ads and news as being tightly linked to the economic sectors heavily contributing to his campaign. Two (non-random) examples: Donald Riegle and John F. Kerry Fergusson (U Andes) Media & Special Interests October 27, 2011 8 / 24
Introduction Donald W. Riegle, Jr (D-MI) One of ”The Keating Five” (Others were Alan Cranston, Dennis DeConcini, John Glenn, John McCain). Riegle’s herfindahl in 1982: 0.592 at the PAC level (sample mean is 0.02); 1 at industry level. Why? He had received money from just a handful of Saving and Loans PACs! In 1988: More than 20 S & L PACs, ˜200 PACs from the Financial sector contributed ˜$500 K to his campaign (larger than any other sector). Fergusson (U Andes) Media & Special Interests October 27, 2011 9 / 24
Introduction Riegle did not run for reelection after the 1989 Savings and Loans scandal: ”I certainly regret and accept responsibility [for actions that] did lend themselves to an appearance of a conflict of interest.“ (1991) Fergusson (U Andes) Media & Special Interests October 27, 2011 10 / 24
Introduction John F. Kerry (D-MA) 1996 high herfindahl and campaign finance scandal Fergusson (U Andes) Media & Special Interests October 27, 2011 11 / 24
Introduction Fergusson (U Andes) Media & Special Interests October 27, 2011 12 / 24
Introduction Fergusson (U Andes) Media & Special Interests October 27, 2011 13 / 24
Introduction Gary Hart (D-CO) The Federal Election Commission has accused a California developer of making contributions exceeding the legal limit of former Senator Gary Hart. The commission contends that Mr. Stein, donated at least $38,750 for the 1980 campaign Fergusson (U Andes) Media & Special Interests October 27, 2011 14 / 24
Introduction More systematic evidence A good proxy Fergusson (U Andes) Media & Special Interests October 27, 2011 15 / 24
Basic Approach Simplest specification Compare in-state versus out-of-state voters response to campaign finance profile. Comparison: in-state versus out-of-state counties ( c ) within each state ( s ) : ⇣ ⌘ ⇣ ⌘ V I in c × h I in c × h O = β in ⇥ h I + β in ⇥ h O (1) ct st st γ s ⇥ P I + θ t ⇥ P I + β 0 X X cst + ε sct where: I =incumbent, O =opponent; V I ct =incumbent share of the two-party vote, in c =1 if c is in-state, 0 if out-of-state; h I st =herfindahl concentration index of special-interest campaign contributions for the incumbent ( h O st , for the opponent); X cst =vector of additional controls (& non-interacted terms). Fergusson (U Andes) Media & Special Interests October 27, 2011 16 / 24
Overview of Results Overview of main results and robustness Voters in in-state media markets punish ”captured” politicians An increase in the concentration of campaign contributions leads to a di ff erential response from voters across di ff erent media markets. Main concern: counties in out-of-state dominated media markets are di ff erent. However: E ff ect IS present: Relying solely on variation in concentration across time within the same 1 county (county fixed e ff ects). Focusing on more comparable in-state counties that are geographically 2 contiguous to out-of-state. Moreover, e ff ect IS NOT present: Moving from in-state towards out-of-state counties without change in media exposure. Fergusson (U Andes) Media & Special Interests October 27, 2011 19 / 24
Overview of Results Nieghboring counties are more comparable Fergusson (U Andes) Media & Special Interests October 27, 2011 21 / 24
Overview of Results Other results Share of connected-PAC money does not a ff ect vote share di ff erentially across counties, unlike concentration. Suggests ”capture” is punished, not just ”money”. Turnout a ff ected when moving from in-state to out-of-state (and not from placebo in-state to out-of-state). Reassuring evidence that it is information. Not significant spillovers to fellow party members, in the same or across states. Main results not driven by di ff erential partisan trends across counties. Media market definition: main results are robust to choice of threshold, or to using a continuos measure of ”in-state” (share of in-state population of the media market). Fergusson (U Andes) Media & Special Interests October 27, 2011 23 / 24
Final Remarks Summing up Voters assess interest group money negatively, but in a relatively sophisticated way: Total interest group money is not punished. Concentration of interest group money from a few sectors is. Consistent with the idea that voters punish “capture” of politicians by interest groups, when media provides the relevant information. Robustness checks suggest results are driven by information, not other omitted variables (like geographical trends), nor by di ff erential partisan trends across counties. Fergusson (U Andes) Media & Special Interests October 27, 2011 24 / 24
Recommend
More recommend
