Sabato's Crystal Ball

So What Just Happened in Virginia?

A brief history of Old Dominion nomination battles

Larry J. Sabato, Kyle Kondik and Geoffrey Skelley, U.Va. Center for Politics May 20th, 2013

Almost all states consistently use primaries to nominate their candidates for statewide office (U.S. Senate, governor, lower statewide elected officials), although there are some exceptions. Utah, for instance, uses a hybrid convention/runoff system, which readers will remember led to an incumbent U.S. senator, Robert Bennett (R), failing to even advance to a two-person primary in 2010. Another state, Connecticut, has a convention that helps set the primary ballot. And South Carolina Republicans recently decided against switching from a primary to a convention for next year’s races.**

Then there’s Virginia: In many contests in recent decades in both parties, nominees for statewide office have been picked at conventions without a primary depending on the preference of the party in that given election. The Old Dominion gives complete latitude to each party in choosing its nomination method every year.

The commonwealth’s curious nomination rules are in the news this week, thanks to last weekend’s Republican Party of Virginia convention in Richmond, where Republican delegates made the highly risky decision to nominate E.W. Jackson (R) for their party’s nomination for lieutenant governor. We’ll let others describe the noxious comments Jackson has made over the years; needless to say, Republican leaders in the state are privately groaning at his nomination. In addition to being next-in-line to the governor, the lieutenant governor, a separately elected position in Virginia, breaks ties in the state Senate, which is currently split 20-20 between the parties (and state Senate seats will not be up for election until 2015). Thus, the election for lieutenant governor is effectively for control of the Senate for the next two years (barring vacancies).

Unquestionably, the convention made Jackson’s nomination possible. The minister and lawyer won about 12,000 votes in 2012’s Republican Senate primary, or 4.7% of votes cast, but because of the tiny turnout at a party convention, he won far fewer votes on the convention’s final ballot than he did in his fringe run for Senate last year. Yet he’s now a statewide nominee who, if elected, would be one heartbeat from the governor’s mansion.

Jackson is far from the first problematic candidate to be selected in a Virginia nominating convention. But a reading of Virginia’s quirky nomination history shows there’s little indication that one method has led to more general election success than the other.

The primary election came to Virginia, like it did in so many states, during the Progressive Era at the start of the 20th century. Touted by Progressive leaders like Robert La Follette (R) of Wisconsin, the primary was billed as a way to empower voters and produce candidates less beholden to the party bosses who dominated nominating conventions. As adopted in Virginia, though, the primary actually helped preserve the power of the conservative Democratic machine (started by Sen. Thomas S. Martin and carried on by the more-famous Sen. Harry F. Byrd, Sr.) that ran the state, because the party conducted the primary in concert with laws that minimized the size of the electorate (poll taxes, literacy tests, etc.). Republicans — a non-factor in Virginia state-level politics until the middle of the century — continued to nominate their candidates in conventions, with the exception of 1949, when a primary experiment led to a divisive, low-turnout election that produced another losing statewide ticket. That same year, a crowded primary field resulted in an anti-Byrd Machine candidate coming perilously close to winning the Democratic gubernatorial nomination. That led to the introduction of a runoff system to protect the machine candidate from losing to an insurgent.

The runoff wasn’t used for another two decades, until 1969, when another multi-candidate Democratic battle led to a runoff between liberal-populist firebrand Henry Howell and the more moderate William Battle, son of former Byrd Machine governor John Battle. Battle narrowly prevailed in the runoff, but the divisive contest helped lead to the election of Linwood Holton, the first post-Reconstruction Republican governor in Virginia history. Holton, like nearly all Republican gubernatorial nominees before him, was nominated in a convention. Four years later, the Democrats didn’t even have a gubernatorial nominee (Howell ran as an independent), and veterans of the collapsing Byrd Machine rallied around Republican nominee Mills Godwin, who had been elected governor as a Democrat just eight years earlier. Godwin beat Howell in a very close contest. (Virginia has a unique law that prevents governors from serving consecutive, four-year terms, but governors can serve non-consecutive terms.)

In 1977, Democrats again had a divisive primary, and they lost their third straight gubernatorial contest. After seeing divisive primaries precede general election defeats in two of three elections, Democrats decided to emulate Republicans by switching to a convention. Democratic conventions selected three straight gubernatorial winners in the 1980s: Chuck Robb (1981), Gerald Baliles (1985) and Doug Wilder (1989). In that last election, Republicans decided to change up their nomination process, and they held their first primary since 1949. The nominee, J. Marshall Coleman, nearly beat Wilder.

The Republicans, having experimented again with a primary, went back to a convention in 1993, and the Democrats stuck with the convention format that year too. That year Republicans nominated a far-right candidate for lieutenant governor, Mike Farris, but Farris’s landslide defeat didn’t prevent George Allen (R) from winning the governorship or Jim Gilmore (R) from winning the attorney general’s office, both also in landslide territory.

Since 1969, Virginians have elected straight-party tickets five times but have split their tickets six times. So voters in this state are perfectly capable of making independent choices for the three top offices. We will bet the 2013 Republican gubernatorial nominee, Attorney General Ken Cuccinelli, is now reminding himself of this history.

Both parties have held nominating primaries and conventions in recent statewide elections. Charts 1 and 2, located at the bottom of this story, list the Old Dominion’s statewide nominees since 1969, and notes how they were nominated and whether or not a candidate won. Note that there’s very little difference in performance between candidates picked in primaries versus those picked in conventions. Parties generally choose a primary or convention based on each year’s unique circumstances and needs.

That’s what happened this cycle. Virginia Republicans were set to nominate their statewide ticket through a primary, and Cuccinelli and Lt. Gov. Bill Bolling (R) were preparing for what could’ve been a competitive primary — although Cuccinelli’s more conservative profile made him an obvious favorite. Yet Cuccinelli and his allies knew that a convention would assure his nomination with far less money expended, and they pushed the state party’s Cuccinelli-controlled central committee to switch to a convention. That decision caused Bolling to exit the race, but it also created an environment that led to Jackson’s nomination. (Democrats are nominating their statewide ticket in a primary this year; gubernatorial nominee Terry McAuliffe is unopposed but there are contested primaries for lieutenant governor and attorney general.)

Republican delegates in Virginia and across the nation are attracted to black conservatives like Jackson, Herman Cain and former Rep. Allen West (R-FL). In a sense, nominating a black conservative demonstrates that GOP activists oppose President Obama on the basis of ideology, not race.

Conventions can also be stampeded by a fiery speech — just think of William Jennings Bryan — and Jackson gave one at the Richmond GOP gathering.

While our history above and chart below focus on statewide executive races, the Jackson nomination is a lot like another nomination decision Republican convention delegates made in a U.S. Senate race a quarter century ago. At the 1988 GOP state convention, there was a three-way race for the Senate nomination to oppose Democrat Chuck Robb. Minister Maurice Dawkins was little-known and was not the frontrunner. But he gave a powerful speech, and Republicans loved the fact that an African-American was espousing strongly conservative views. Dawkins won the nomination in an upset, but he went on to lose to Robb by more than two-to-one.

The Jackson problem can be managed by the GOP, but they need to be sure-footed about it. There’s always the potential for disaster. It’s an unwelcome, major distraction for Cuccinelli.

The old Chinese proverb applies: Be careful what you wish for. Cuccinelli and his backers insisted on a convention; they now must live with the consequences.

Chart 1: Republican nominees for Virginia statewide office, 1969-2013

Chart 2: Democratic nominees for Virginia statewide office, 1969-2013

Notes: Bolded candidates won the general election. *Indicates candidate was unopposed for nomination at either the convention or in a primary race. #Battle won the 1969 gubernatorial nomination after a runoff. ^The independent city Nansemond no longer exists; it is now part of the city of Suffolk. In 1973, independent Henry Howell was the de facto Democratic nominee, though he received only a “commendation” from state Democrats and was listed on the ballot as an Independent.

**The Green Papers has an excellent state-by-state rundown of how candidates are nominated.

You’re Not From Around Here, Are You?

How migration does — or doesn't — change how a state votes

Geoffrey Skelley, Political Analyst, U.Va. Center for Politics May 16th, 2013

“I was born an American; I will live an American; I shall die an American.” — Daniel Webster

While Daniel Webster died an American in 1852, his political legacy does not belong to just one state, but two: New Hampshire and Massachusetts. Born in New Hampshire, Webster represented the Granite State in the House of Representatives from 1813 to 1817. But he then moved to Massachusetts seeking to improve his legal career, only to wind up returning to the House as a Bay State congressman in 1823. (Republican ex-Sen. Scott Brown is currently pondering the reverse move.) Webster went on to have a lengthy stay in the Senate, becoming part of the upper chamber’s revered “Great Triumvirate” with Henry Clay and John C. Calhoun. As a transplanted New Hampshirite representing Massachusetts, Webster’s individual case demonstrates how politics can be affected by the movement of Americans from state to state.

In the aftermath of the 2012 election, the “demographics as destiny” discussion has dominated political analysis, with the latest data being provided by last week’s U.S. Census report on the 2012 electorate. But one demographic statistic hasn’t received much attention in the conversation: state nativity rates — that is, the percentage of people residing in a state who were born there. Does that statistic tell us anything about the politics of a state?

If we order the states by nativity percentage (Chart 1) while also considering which party each state supported in 2012, we find that there are more Blue states than Red states with lower levels of nativity. Yet it’s obvious that high nativity rates in Iowa, Michigan, Ohio, Pennsylvania and Wisconsin did not keep Barack Obama from winning those states in 2012.

Chart 1: Percentage of people residing in a state who were born there in 2010, shaded by 2012 vote

Source: 2010 Census data

If we dig deeper by analyzing the relationship between state nativity rates and President Obama’s support level, we find evidence that, superficially, nativity rates tell us little. As Chart 2 indicates, there is only a weak negative correlation (R = -.235) between a state’s nativity percentage and the percentage of the vote Obama received in the 50 states plus Washington, D.C. The analysis also tells us that nativity rates explain very little of the variation in Obama’s performance from state to state. In other words, a state with a low percentage of native-born residents was not clearly more likely to support the president’s reelection bid.

Chart 2: Simple linear regression analysis of 2012 Obama vote and state nativity rates

Nonetheless, on a case-by-case basis, the nativity rate can tip us off to important changes within a state. While it doesn’t tell us much about Nevada, for example — which has always been a state with a high number of residents born elsewhere — it can reveal something about states where there have been significant decreases in nativity over time.

Appropriately for our purposes, the Crystal Ball’s home state of Virginia exemplifies how nativity can matter as a part of the demographics discussion. Chart 3 details the nativity percentage in each state over the last century (from 1910 to 2010) based on U.S. Census data. Many states have had relatively stable nativity rates: for instance, perennially competitive Ohio and Pennsylvania have consistently had populations where roughly three of every four residents were native-born over the past 100 years. However, relatively sharp drops in nativity explain a lot in a few states, including Virginia, which has gone from having one of the nation’s highest nativity rates in the 1910 census to having one of the lowest in the 2010 census.

Chart 3: Percentage of people residing in a state who were born there, 1910-2010

Source: Decennial census data, 1910-2010. Click here to download data.

Note: While Arizona and New Mexico did not become states until 1912, the Census did have nativity data from 1910 available for them.

In fact, over the last century, Virginia saw the largest decrease in the percentage of people living in the state that were born there (as well as the sharpest downward slope). For the first time in the modern era, the 2010 Census recorded that a slight majority of Virginia residents were born outside the state. That should not come as a surprise: People from across the country and around the world have been moving to the D.C. suburbs to work in government, government relations and corporate jobs in or near the nation’s political center for decades, while others have taken positions in the service industries that have boomed while the region has grown. Consequently, the economic engine of Northern Virginia has altered the political culture of the state. As a Public Policy Polling survey showed in August 2012, those who had lived in the state 30 years or fewer favored Obama (particularly the 10 years or fewer group), while those who had lived in the state longer than 30 years favored Mitt Romney. Although PPP’s poll found equal support for Obama among native and non-native Virginians, the “length of stay” factor in Virginia shows how younger native Virginians — a more diverse group — were more inclined to support Obama. The poll also showed that the Virginia electorate was half native, half non-native, matching the general population as reported by the Census figures.

However, Virginia’s nativity rate had already shrunk significantly before the state went Democratic in 2008, the first time it had done so since 1964. An additional factor within the falling nativity rate is the increasing diversity and background of people moving to the state. The foreign-born percentage of the state’s population grew from 5.0% in 1990 to 8.1% in 2000 to 11.4% in 2010. As of 2011, 47% of those who were foreign born in Virginia had become naturalized citizens. And this doesn’t account for any American-born children they or non-citizens have had in the last two decades who have reached voting age.

Nationally, the groups most likely to move are minorities, and while this was true previously, the sheer numbers have increased, particularly among Asian and Latino groups. As of 2011, the top three countries of birth for foreign-born Virginians were El Salvador, India and South Korea. In 2000, about 8% of the state’s population was neither solely white or black; in 2010, that figure stood at 12%. (Although the Census Bureau has not previously included Hispanic as a racial category, about 47% of Virginia Latinos counted themselves as something other than white or black in the 2010 Census.) Connecting the dots, simply look at the 2012 exit poll in Virginia: roughly two-thirds of Asian-Americans and Hispanics voted for Obama. While black voters obviously played a key role in Obama’s success in the Old Dominion (20% of the state’s electorate), voters who did not consider themselves white or black were 10% of the state’s turnout in 2012.

The Virginia example shows how an increasingly diverse population (whose most varied members are the most likely to move around) can impact an electorate and potentially alter a state’s political culture. It’s also important to consider how nativity rates could signal changes in other states. Returning to Chart 1, the fact that a number of Midwestern Obama states (plus Pennsylvania) have high nativity rates could add to our previous discussion of how the very white Midwest may be key to future presidential successes for the GOP.

Conversely, some Southern states, notably North Carolina and Georgia, exhibit similar nativity patterns to Virginia. Like Virginia, North Carolina has become more politically competitive in presidential races as population shifts have occurred, with people from all regions moving to the Research Triangle and Charlotte areas. But while Virginia’s 2012 electorate was about 50% native, polling found during 2012 that the Tar Heel State’s electorate was slightly more than 60% native. Georgia, too, could conceivably be headed for a future of greater electoral competitiveness as its population becomes more diverse and less home-grown.

Although Daniel Webster’s time saw the rise of Manifest Destiny, the pull of the California Gold Rush and famous treks on the Oregon Trail, Americans are exceedingly more mobile today than they were in the first half of the 19th century. And as Virginia demonstrates, who stays and who goes can have major political consequences.

Sanford Joins “the Underachievers”

Kyle Kondik, House Editor May 9th, 2013

In winning his special election victory on Tuesday night, incoming Rep. Mark Sanford (R-SC) joined a dubious but sizable bipartisan House caucus: The Underachievers.

Many House observers — including the Crystal Ball — have focused, understandably, on the small number of House members elected from districts won by the other party’s presidential nominee. These represent obvious targets for the other party, although there are only a handful of these districts (nine Democrats reside in seats won by Mitt Romney, and 17 Republicans occupy seats won by President Obama). So in an effort to expand their maps, strategists from both parties will also look at The Underachievers, who ran behind their party’s presidential nominee in their districts. Sanford is in the club because he won about 54% of the vote in the special election, but Mitt Romney won about 58% in the district in last November’s presidential election.

Close to three of every 10 House representatives are members of The Underachievers. These 60 Democrats and 63 Republicans could be vulnerable in a primary or general election, but there are plenty of extenuating circumstances that explain their underperformance, too.

Chart 1: House Democrats who ran behind President Obama in their districts

Chart 2: House Republicans who ran behind Mitt Romney in their districts

Notes: *Represents members who ran against members of their own party in the fall general election (applies to some races in California and the 3rd District runoff in Louisiana; in the LA-3 race, the runoff results were used). ^Represents members elected in 2013 special elections. Names in bold are freshmen members; new members with previous service time, like Sanford, are counted here as freshmen.

Source: Election results from state-level sources; 2012 district-level presidential results from Daily Kos Elections.

The most obvious and common reason for a member to be on this list is because he or she is a freshman: 85 new members were elected either last November or in a special election since the November election (Sanford and Democratic Rep. Robin Kelly of IL-2), and of those more than half — 45 members (23 Republicans and 22 Democrats) — underperformed their party’s presidential nominee. That makes sense; it’s harder to run as a challenger than as an incumbent. As they become better known in their districts, many of these members will improve their performance over time (which would be unsurprising in an institution where on average more than nine of every 10 members who run for another term get it).

Some of the biggest underachievers are a number of California Democrats, but there’s a big caveat. The Golden State instituted a new election system in 2012: All candidates from all parties run against each other in the same primary, and the top two finishers advance to the general election. So in some instances — marked with asterisks on the charts above — two members of the same party face off in the general election. That explains why some Democratic members severely underperformed President Obama — Rep. Janice Hahn (D, CA-44) is the biggest “underachiever” on this list, but she ran against a fellow Democrat, ex-Rep. Laura Richardson, last November. Hahn, a white woman representing a district that is only about 7% white, will probably continue to face primary challenges in her district, which because of California’s rules will often extend into the general election given how Democratic the district is (Obama won 84.7% there).

Underperformance can be a sign of past or future primary trouble for others, too. For instance, Rep. Scott DesJarlais (R, TN-4) ran almost 10 points behind Mitt Romney in 2012, but DesJarlais’ embarrassing personal history explains why he lagged. If DesJarlais somehow wins his 2014 primary, Democrats might have a shot at him, but that’s only owing to his own personal weakness, not because his blood-red district (65.3% for Romney) has any interest in electing a Democrat. Some other underperformers on this list faced primary trouble last year — like Reps. Spencer Bachus (R, AL-6) and Charlie Rangel (D, NY-13) — which could have also affected their general elections (although they still won by huge margins).

Several potential Senate candidates are included on this list. The biggest name that jumps out is that of Rep. Colleen Hanabusa (D, HI-1). If the handful of members from California who ran against members of their own party in the general election is excluded, no current House member ran further behind his or her party’s presidential nominee than Hanabusa did in her Honolulu-based district in 2012. In Hanabusa’s defense, President Obama has a special appeal in Hawaii, which makes the Aloha State seem more Democratic at the presidential level now than it usually is. For instance, Hanabusa’s current district — which changed very little in post-2010 redistricting — only gave John Kerry a 53%-47% victory over George W. Bush in the 2004 presidential race, as opposed to the sky-high 69.7% of the vote it gave to Obama last November. Still, these results lead one to question Hanabusa’s appeal as she seeks to unseat appointed Sen. Brian Schatz (D-HI) in a primary. Other potential Senate candidates who ran behind their presidential nominee in 2012 (though by much, much smaller margins than Hanabusa) are Reps. Kristi Noem (R, SD-AL), Steve Daines (R, MT-AL), Tom Cotton (R, AR-4) and Justin Amash (R, MI-3). All four are newer members of the House, and Cotton and Daines are freshmen. In other words, these results don’t tell us much about how these representatives might perform if they tried to move up to the Senate.

Redistricting explains some of the underperformers. Six of the nine Republicans who now represent North Carolina in the House ran behind Romney, and Republicans heavily redistricted that state in order to pick up seats there (they largely did, winning three Democratic-held seats and nearly winning a fourth). Three of the six Republican Tar Heel State underperformers are first-time representatives; Rep. Richard Hudson (R, NC-8) ran about five points behind Romney, but that’s because he defeated a Democratic incumbent, ex-Rep. Larry Kissell (D). In all likelihood, 2012 will probably be the hardest race he faces under the current map. On the other side of the ledger, the Democratic redraw of Illinois allowed Democrats to grab several seats there, but many of their winners lagged behind favorite son Obama. Notably, Rep. Bill Foster (D, IL-11) actually ran ahead of Obama, yet his district seems to be receiving outsized attention from Republicans, while there’s hardly a peep of anyone trying to run against Rep. Tammy Duckworth (D, IL-8), the war hero who nonetheless lagged behind Obama in her district (and ran against a particularly weak Republican incumbent, loudmouth ex-Rep. Joe Walsh).

National Democrats and Republicans sifting through this list can find some targets, although some are more attractive than others. Democrats, for instance, could try to dislodge freshmen Reps. Keith Rothfus (R, PA-12) and Andy Barr (R, KY-6), who each defeated solid Democratic incumbents in strong Republican presidential districts. But Rep. Jackie Walorski (R, IN-2) might be more vulnerable. She’s a freshman too, but unlike Rothfus and Barr, she did not face an incumbent in 2012, and she lagged further behind Romney than all but two Republican members of the House: the aforementioned DesJarlais, and Rep. Markwayne Mullin (R, OK-2), who won an ancestrally Democratic seat in 2012 (only two Republicans had ever been elected to it prior to Mullin).

Two of the worst Democratic underperformers on this list, Reps. David Cicilline (D, RI-1) and John Tierney (D, MA-6), had unique problems in 2012 — Tierney’s family was embroiled in a criminal case, and Cicilline faced questions over his stewardship of Providence while mayor. Will these problems blow over, or will these two New England Democrats continue to underperform?

In many of these races, significant third party challengers siphoned votes from one or both candidates. For instance, Rep. Dan Maffei (D, NY-24) ran about eight points behind Obama, but a Green Party candidate also won close to 8% of the vote in the race. Presumably, that candidate siphoned more votes from  Maffei than from his Republican opponent, ex-Rep. Ann Marie Buerkle.

Nonetheless, Maffei was one of seven Underachievers who not only lagged behind their party’s presidential nominee, but also won less than 50% of the vote in their districts. The other Democrats are the aforementioned Tierney (MA-6), along with Reps. Carol Shea-Porter (NH-1) and Kyrsten Sinema (AZ-9); the Republicans were Walorski (IN-2) as well as Reps. Dan Benishek (MI-1) and Rodney Davis (IL-13). All could face stiff challenges.

Meanwhile, Sanford’s days of being challenged by Democrats could very well be over, though his membership in The Underachievers is but one reason why his next primary could be his last.