Lot-et-Garonne 3rd (France) by-election: Losing streak
A legislative by-election was held in the Lot-et-Garonne’s 3rd constituency in France on June 16 and 23, 2013. The constituency’s deputy, Jérôme Cahuzac (Socialist Party, PS), was compelled to resign his seat on April 16, 2013 after having been removed from Prime Minister Jean-Marc Ayrault’s government on March 19, 2013. In France’s semi-presidential system, a sitting deputy relinquishes his legislative seat to his suppléant (running mate) once he enters government. Until 2008, upon leaving the government, the former deputy needed to run and win in a by-election to retrieve his seat in the National Assembly. Since a constitutional reform in 2008, the former deputy automatically regains his seats upon leaving the government. However, given the nature of the accusations against him which had forced him to leave the government, Cahuzac had little choice but to resign his seat as deputy, creating a vacancy.
French legislative elections or by-elections are fought on a two-round system. A candidate must win over 50% of valid votes representing at least 25% of registered voters to win outright by the first round. If a second round is organized, all candidates who have won over 12.5% of registered voters are qualified for the runoff; or, if no candidates meet this requirement, the top two candidates in the first round. In by-elections were turnout is almost always lows, this means that only the top two candidates will qualify.
Jérôme Cahuzac held this constituency between 1997 and 2002 and again between 2007 and 2012. Cahuzac gained a profile as a moderate and ‘expert’ on fiscal and budgetary issues, serving as president of the National Assembly’s finance commission after 2010. After having served in François Hollande’s campaign team in the 2012 presidential election, he was name junior minister for the budget in Jean-Marc Ayrault’s government in May 2012, after Hollande’s victory. With a reputation as an orthodox ‘budget hawk’, he was in charge of dealing with France’s large government deficit and public debt.
In December 2012, the online newspaper Mediapart accused Cahuzac of tax evasion, alleging that he had a hidden offshore account in Switzerland until 2010 (at which point it was closed and the money transferred to another account, in Singapore). Unwilling to reveal their sources, Cahuzac vehemently denied Mediapart‘s allegations and received the backing of the President and Prime Minister. At that point, Mediapart was running on the basis of a voice recording from 2000, a conversation between Cahuzac and his asset manager. This recording’s veracity was confirmed by sources close to Cahuzac and some of his former local political rivals. On March 19, 2013, a Parisian court opened a preliminary inquiry into suspicions of tax evasion. While continuing to claim his innocence, Cahuzac was removed from government that same day. On April 2, Cahuzac was indicted on a charge “laundering of tax fraud proceeds and money laundering of proceeds from a company whose products or services are covered by Social Security” and he was forced to admit, on his blog, that he did indeed hold 600,000 euros (likely more) in an offshore account. Swiss authorities had already discovered a bank account belonging to him.
President Hollande and Prime Minister Ayrault minced no words in disowning him, Hollande saying it was a “moral fault”. On April 9, Cahuzac was excluded from the PS. However, legitimate suspicions abound as to whether or not Hollande, Ayrault and the top echelons of the PS knew that Cahuzac had concealed his offshore account before he admitted to it himself. Mediapart claimed that the interior ministry authenticated the voice recording in a three-page report to the presidency; it also claimed that Cahuzac’s then-senior minister, Pierre Moscovici (minister of the economy and finances) intervened in January 2013 to protect Cahuzac.
The Cahuzac affair, the first major scandal for the new Socialist government, could not have come at a worst time for the government. A bit over a year after taking office, President Hollande’s disapproval rating stands at 70% (!) and his Prime Minister’s disapproval ratings only slightly better at 65-67%. This is the lowest approval rating for any President after a year in office, and approaching all-time lows for presidential unpopularity (mid to low 20s).
A large part of this unpopularity stems from the politcal and economic conjuncture. France is wracked with high unemployment (over 10%), an economy in recession and a very large public debt (90%). At least some of France’s economic woes are beyond the government’s control, although voters will invariably lay the blame on a poor economy on whoever has the bad luck of being in power. It is likely that if President Nicolas Sarkozy had won reelection in May 2012, his approval ratings would be just as low as Hollande’s. However, the economic crisis has only aggravated matters and a good part of this government’s unpopularity is of its own making.
At times, the government has been a bit like a deer in the headlights when it comes to dealing with the economic crisis. It has been seen as powerless, lost and incompetent in its handling of the economy. The right has criticized the ‘amateurism’ and jumbled response of the government and denounced high taxes. Many on the left, however, also dislike the government. Hollande and the PS won the 2012 election on a fairly anti-austerity platform full with flowery rhetoric about ‘growth’ and nice things, but once in power it has largely continued Sarkozy’s austerity policies (disguised as ‘efforts’ because austerity is unpronounceable by governments since the 1980s). Hollande approved the European Fiscal Compact without any substantial changes, despite having pledged to renegotiate it. His government has implemented harsh austerity measures, including tax increases and spending/job cuts in the public sector. The Constitutional Council has also forced him to scrap, entirely, his much-publicized 75% tax on incomes over a million euros. With good reason, many on the left feel that Hollande’s policies are no different than his predecessor’s policies.
Other election promises – constitutional reforms, cracking down on dual office holding (cumul des mandats) and so forth – have been watered down or indefinitely delayed. The government was successful in passing its landmark same-sex marriage and adoption law in May 2013, but it was passed at the price of riling up social conservative and Catholic public opinion in the form of enormous anti-gay marriage rallies.
On the symbolic aspect of things, Hollande had made a big deal of Sarkozy’s centralizing, autocratic and flashy presidential style and he famously presented himself as the ‘normal President’ in contrast to the ‘hyper-President’. Yet, the symbolic changes at that level have been slow to come. The ‘normal president’ mantra was quickly dropped. By choosing his close ally Jean-Marc Ayrault as Prime Minister, Hollande signaled that he was continuing in Sarkozy’s, rather than Mitterrand’s, footsteps by choosing a close ally and partner as Prime Minister. While the left criticized Sarkozy for sidelining the Prime Minister and concentrating powers in the executive branch, Hollande has done largely the same. Ayrault, a year later, appears effaced and a mere ‘sidekick’ in comparison to his President.
Within the government, there has often been cacophony and public disagreements between cabinet ministers, which Ayrault has struggled to deal with. Cabinet solidarity appears to be quite shaky. For example, Arnaud Montebourg – the minister of industry and a leader of the PS’ left-wing faction – told Ayrault that he was managing France like the municipal council in Nantes (Ayrault was mayor of Nantes before becoming Prime Minister) and that he was ‘pissing off’ everybody with the controversial new airport project on the outskirts of Nantes (which Ayrault strongly supports, along with most of the PS, but not Montebourg and the Greens). Ayrault confirmed Montebourg’s insubordination but he was not fired. There have also been internal disagreements between the PS and its most demanding junior partner, the Greens (EELV) – which is seriously considering leaving the governing coalition. The Left Front (FG) led by Jean-Luc Mélenchon and the Communist Party (PCF) have been vocal critics of government policies.
There have been eight legislative by-elections since the last legislative elections (seven of them actually ‘count’ because Wallis-et-Futuna has no impact on national politics and vice-versa – the ‘right-winger’ backed by the UMP who won that by-election now sits with the PS…). The PS held four of these eight seats, and it has lost all of them. It was eliminated by the first round in two seats it already held and in two other seats in which it was the main challenger in 2012.
The first ‘mid-term’ electoral test for the government will be municipal elections in March 2014, followed by European elections in June 2014. The left might manage to hold on fairly well in the municipal elections, but European elections are usually brutal for the governing party and the PS will likely take a massive thumping. Many assume that Hollande will change Prime Ministers after the Euros in June 2014.
Lot-et-Garonne’s third constituency includes the northeastern region of the Lot-et-Garonne, the region centered around Villeneuve-sur-Lot. The constituency includes the cantons of Beauville, Cancon, Castillonnès, Fumel, Laroque-Timbaut, Monclar, Monflanquin, Penne-d’Agenais, Prayssas, Sainte-Livrade-sur-Lot, Tournon-d’Agenais, Villeneuve-sur-Lot Nord, Villeneuve-sur-Lot Sud and Villeréal. Its boundaries have remained unchanged since 1986.
This is a predominantly small town constituency, with some more rural and sparsely populated regions. Villeneuve-sur-Lot, with a population of 23,530 is the largest city in the constituency. Sainte-Livrade-sur-Lot (pop. 6,410) and Fumel (pop. 5,186) are the two other major towns in the constituency, although a good number of towns have over 2,000 inhabitants. In general, the main centres of population are concentrated along the Lot river, which flows through the constituency and its three largest communes. The ‘inland’ communes tend to be rural, isolated from major urban centres. Historically an agricultural region, the constituency nowadays tends to be lower middle-class, with a large blue-collar (ouvriers and employees) population and a high percentage (36%) of retirees.
Historically, the department (and this constituency) was rural and agricultural – wheat, wines, fruits and vegetables being the dominant crops in the department. As recently as 1968, agriculteurs (farmers who owned and work their land) still made up a plurality of the working population in most of the constituency, excluding the major cities and the Fumélois. The Lot-et-Garonne’s social structure was a mix of smallholdings and métayage (a form of sharecropping), although métayage was more dominant in the Marmandais (the western half of the department). Agriculture declined significantly after the Second World War, continuing a rural exodus which had begun in the mid-nineteenth century and continued into the 1970s, although urban areas grew considerably after the 1920s.
The exception to the agricultural nature of the constituency was found in the three main urban centres. Villeneuve-sur-Lot has always served as an important commercial centre by virtue of its geographic location on the Lot river. Sainte-Livrade-sur-Lot’s economy has been more closely tied to its rural surroundings. Fumel and its immediate surroundings, located at the eastern extremity of the constituency along the Lot river, has had a large iron working/metallurgical industry since the 1850s. Some rural areas, however, have some industrial backbone as well – construction, small businesses and so forth.
After the Algerian War and Algerian independence (1962), a large number (about 20,000) of pieds noirs and harki refugees settled in the Lot and Garonne valleys. In the interwar years, the region had already received Italian and Spanish immigrants.
Politically, this region of the department has been a closely disputed bellwether since the 1980s. The Communist Party (PCF), which was extremely powerful in the Marmandais as early as the mid-1920s until 1981, was never as strong in the Villeneuvois, although it did fairly well in the industrialized Fumélois and in some rural cantons in the north. Instead, the constituency remained dominated by notables and moderate ‘bourgeois parties’ – Georges Leygues, a centre-right republican who served President of the Council between September 1920 and January 1921, held the continuously seat between 1885 and 1933. The Radical Party, the epitome of the parti de notables which dominated small-town republican and anti-clerical areas such as this one for decades, was rather strong in the region as well. In 1958, a former Radical-turned-Gaullist (Jacques Raphaël-Leygues, none other than Georges Leygues’ grandson) won this seat for the Gaullist UNR. He was defeated in 1962 by Édouard Schloesing, a moderate Radical who refused the Programme commun with the PS and PCF in 1972 and was reelected – one last time – in 1973 for the centrist ‘Reformist Movement’ (MR). The first left-winger to represent the constituency was Marcel Garrouste, a Socialist elected in 1978, 1981 and 1988.
Nevertheless, the left – echoing a long Radical tradition which retained leftist overtones for quite some time – dominated presidential politics until 1981. Mitterrand easily won the constituency in 1965, 1974 and 1981 – on the current boundaries, he won 55.8% of the vote in 1981. The substantial shift away from the left came in one shot – in 1988, Mitterrand actually performed below his national average in the seat, winning only 52.8% in the second round. Since then, numbers have been stabilized – the constituency is a pure bellwether. The results in presidential elections since 1995 have been remarkably close to national numbers: 53.2% for Chirac in 1995, 53.7% for Sarkozy in 2007 and 51.8% for Hollande in 2012. Even in the first rounds, with the exception of the far-right which tends to be a few points above average and the Greens and far-left who are a few points below, the numbers for the PCF, PS, UDF and RPR-UMP have been very similar to national numbers.
What explains the sudden shift away from the left, between 1981 and 1988? The left-wing anti-clerical and republican tradition which had prevailed for over a hundred years declined, with a rural exodus, urbanization and social dislocation bred by such changes. This was likely aggravated by the economic crisis of the 1980s. Immigration became a major issue in this region starting around the same time. The region’s strong fruit and vegetable industry has always required a large seasonal workforce. While these roles were often filled by Italians, Spaniards or Portuguese in the 1960s and 1970s, they were progressively replaced by Moroccan and other North African immigrants. The constituency has a fairly large foreign population (6%), although some of those ‘foreign nationals’ are British or other EU citizens who settled in rural southern France.
Each main party’s strength is almost evenly distributed throughout the constituency. In the 2012 runoff, Hollande did best in the cantons of Fumel (59.6%) and Tournon-d’Agenais (57.7%) – both industrialized (metal) areas in the Fumélois; but in all other cantons, his support ranged from 47% to 54%. Similarly, Marine Le Pen’s support was fairly evenly spread – she did not do significantly worse in major cities (she won 23% in Sainte-Livrade-sur-Lot, although she performed a bit below average in the other towns) and her support ranged from 19% to 23%. The FN traditionally tends to be strong along the Lot and Garonne rivers, a region mixing Pieds-Noirs with seasonal immigration, fruit/vegetable farms and small businesses (a perfect recipe for a strong FN vote); but in 2012, Marine Le Pen improved on past far-right performances in the more rural ‘inland’ cantons.
Villeneuve-sur-Lot, like many similar commercial and rather bourgeois towns in the old Radical southwest, was a Gaullist stronghold until 2001 – in many elections (such as 1981), the right’s strength was concentrated in Villeneuve-sur-Lot while the left prevailed in the surrounding rural areas. It has since shifted to the left. In 1998, Jérôme Cahuzac gained the canton of Villeneuve-sur-Lot-Sud and he gained the town hall from the RPR’s Michel Gonelle in a triangulaire in 2001. He was reelected by the first round in 2008. Hollande narrowly won Villeneuve-sur-Lot in the 2012 runoff.
As a rural area, the Chasse, pêche, nature et traditions (CPNT) did very well in the constituency in the 1990s. It won 10.1% in the 1999 European elections, and CPNT presidential candidate Jean Saint-Josse did rather well in isolated rural cantons in 2002 – up to 19% support.
The PS narrowly held the seat in the 1988 legislative elections, with former deputy Marcel Garrouste. However, Garrouste retired prior to the catastrophic 1993 elections. Hurt by the atypical candidacy of Anne Carpentier, owner of a satirical local paper, the PS found itself eliminated by the first round and the second round was a fraternal runoff between the UDF mayor of Monflanquin, Daniel Soulage, and the RPR mayor of Villeneuve-sur-Lot, Michel Gonelle. Soulage won a by a hair. In 1997, however, the PS – with Jérôme Cahuzac – staged its revenge and narrowly defeated Soulage (UDF) in the runoff, with 50.7% for Cahuzac. In 2002, the pendulum swung back to the right – Alain Merly, the UDF-UMP mayor of Prayssas, defeated Cahuzac in the runoff, with 51.8%. Merly served only one term, retiring before the 2007 election. Cahuzac, trying to regain his old seat, faced Jean-Louis Bruguière, a famous counter-terrorism magistrate. Cahuzac defeated Bruguière in the second round, winning 52.1%.
The 2012 election was different from all others. Jérôme Cahuzac, who had just been named to cabinet and was riding on wave of notoriety (and popularity), won a record 46.9% by the first round and steamrolled the UMP’s candidate, Jean-Louis Costes, in the second round with 61.5%. Cahuzac likely received a substantial local boost from his nomination to cabinet, a ‘cabinet effect’ which benefited a few other of his (former) colleagues in June 2012.
No less than seventeen candidates faced off in the first round of the by-election.
The UMP candidate, as in June 2012, was Jean-Louis Costes, the mayor and general councillor for (the strongly left-wing) Fumel and the leader of the departmental opposition in the general council. Jean-Louis Costes appears to be on his party’s right – he is a close sympathizer of the Mouvement initiative et liberté (MIL), a hardright Gaullist movement known for its strongly anti-leftist, anti-immigration and anti-Islam attitudes. Its ranks include the likes of Bernard Debré, Yves Guéna, Charles Pasqua and Jean Tiberi. He faced some minor right-wing dissidents, including Joffrey Raphaël-Leygues, the 18-year old grandson of former UNR deputy Jacques Raphaël-Leygues (and great-great-grandson of Georges Leygues), who ran as an ‘extreme centrist’ (I really love that).
The PS candidate was Bernard Barral, a retired 66-year old businessman. The left was shaken up for a few weeks in early-mid May by persistent rumours that Jérôme Cahuzac would seek to regain his old seat as a PS dissident candidate. Cahuzac has taken his ‘forced’ resignation and subsequent lynching by his former colleagues quite badly – while he has admitted that he did have an offshore account, he doesn’t seem to think that it was a big deal and he feels that he has been betrayed by his old party and colleagues. He seems quite intent on taking his revenge, and he is out for blood – particularly Socialist blood. In early May, he apparently surveyed the ground for a potential dissident candidacy and said – with such chutzpah – that “some are speaking for me without a mandate to do so”. A poll by LH2 showed that he would have won 11% as a dissident candidate. He ultimately decided against running, but it’s quite clear that he intends to stage a comeback – perhaps in the 2014 local elections in Villeneuve-sur-Lot. The PS also faced FG and EELV candidates.
The FN candidate was 23-year old Etienne Bousquet-Cassagne, the departmental secretary (leader) of the FN in the department. Bousquet-Cassagne symbolizes a new generation of FN candidates being promoted by the party’s leader, Marine Le Pen. Seeking to assert her control over the party, sideline the ‘old guard’ and sanitize the far-right’s image (dédiabolisation), she has been promoting a new generation of young leaders and candidates to prominent roles in the FN’s leadership (Florian Philippot, a 31-year old Gaullist technocrat, is the FN’s vice-president and rising star). This strategy has ruffled a lot of feathers and displeased the more radical (neo-fascist) old guard of the FN, but it is quite irrelevant in the electoral scheme of things – FN dissidents have invariably been crushed and killed off by the FN leadership since 1998.
There was also a slew of perennial candidates – most of whom didn’t actually live in the department, let alone the constituency. Nicolas Miguet, leader of the anti-tax ‘Rally of French Taxpayers’ (RCF) and tax fraudster, ran here. You also had a monarchist candidate, a libertarian, one for the far-left (NPA), a Pirate and a few other jokers.
Turnout was 45.88%, down from 64.16% in 2007 but nonetheless an excellent turnout for a by-election.
Jean-Louis Costes (UMP) 28.71% (+1.71%)
Etienne Bousquet-Cassagne (FN) 26.04% (+10.33%)
Bernard Barral (PS) 23.69% (-23.17%)
Marie-Hélène Loiseau (FG) 5.08% (+0.58%)
Anne Carpentier (Parti d’en rire) 3.28%
Lionel Feuillas (EELV) 2.78% (+0.75%)
Yamina Kichi (MoDem) 2.33%
Benoît Frison-Roche (DVD) 2.32%
Hervé Lebreton (SE) 1.69%
Joffrey-Raphaël Leygues (DVC) 1.44%
Maria-Fé Garay (NPA) 1.11% (+0.46%)
François Asselineau (UPR) 0.58%
Nicolas Miguet (RCF) 0.42%
Cédric Levieux (Pirate) 0.19%
Stéphane Geyres (Libertarian) 0.17%
Michel Garcia-Luna (AR) 0.16%
Rachid Nekkaz (RSD) 0.00% (0 votes!)
Turnout in the second round was 52.47%. Blancs et nuls votes (invalid) stood at 7.48%, up from 2.18% in the first round.
Jean-Louis Costes (UMP) 53.76%
Etienne Bousquet-Cassagne (FN) 46.24%
The PS candidate was eliminated by the first round, similar to what happened in March in the Oise’s 2nd constituency. In this case, however, it is even worse. The Oise result was bad for the PS, no doubt about that, but it was a right-wing constituency (and one which has been moving rightwards consistently since 1981) – Sarkozy won 56% in the runoff there and Hollande had placed third in the first round with only 22% of the vote. This constituency, however, is a bellwether constituency (not a left-wing stronghold as the 2012 results indicate). Of the seven ‘normal’ by-elections since June 2012 (again, excluding idiosyncratic Wallis-et-Futuna), this is the first one in a seat which Hollande won in the runoff.
After losing in the Oise, the PS lost two by-elections in constituencies for French citizens living abroad. The PS did, all things considered, fairly well in the first constituency (United States and Canada), winning 25% (with an additional 7.4% for the EELV candidate, who had backed the PS in 2012) and saving face (despite losing) in the second round with 46.8%. But that was probably due, in large part, to the flukes of low turnout (13%) and the personality of the UMP’s candidate, Frédéric Lefebvre – a former junior cabinet minister under Sarkozy who had been parachuted into the constituency in June 2012, much to the distaste of the local right-wing networks. Although Frédéric Lefebvre managed to win the seat in the by-election, his result was nothing to write home about.
However, the result for the PS in the eight constituency (Italy, Greece, Turkey, Israel) was disastrous – in a seat which it had won in 2012 (but that was a huge fluke), it was eliminated by the first round with only 14.6% of the vote. But while that was a bad result, no doubt about it, the PS’ victory in June 2012 owed a lot to the fraternal (Israel-induced) divisions of the right. Hollande had won 37% in that constituency in the May 2012 runoff, with Sarkozy handily winning on the back of quasi-unanimous support (around 90%) in Israel. The UMP candidate in June 2012, who ran again in the by-election, was accused by a right-wing dissident of being too pro-Palestinian (even if she was quite pro-Israeli, just not to the extent of the dissident in question), and the dissident went on to back the PS in the second round. In the by-election, the PS was hurt by a EELV candidate who drew a lot of left-wing/PS voters away in Greece, and it was eliminated by the first round – the runoff was a fraternal battle between the UMP’s Valérie Hoffenberg and Meyer Habib, running for Jean-Louis Borloo’s ostensibly centre-right Union of Democrats and Independents (UDI). Meyer Habib, who was publicly endorsed by Israeli Prime Minister Benjamin Netanyahu, won the second round with 53.4%.
The PS’ result is beyond disastrous. To be fair, commentary on the loss of nearly half of their votes from June 2012 (or -23.2% in percentage terms) should be tempered by the observation that the PS’ result in June 2012 was abnormally high given the constituency’s swingy nature – a lot of those votes were likely fairly non-ideological personal votes for Cahuzac, who, back then, was something of a rising star and quite popular both in his constituency and France as a whole. Cahuzac’s 46.9% in the first round in June 2012 should not be interpreted as a sign that this is a solidly left-wing seat or that 46% represents the average share for the PS. Nevertheless, even when compared to more ‘normal’ results – 2007, 2002 and 1997 – it is clear that the PS has taken quite a tumble. In 2007, Cahuzac won 37.6% in the first round. In 2002, he had won 34% and in 1997 he had taken 27.6% (his gains between 1997 and 2002 likely came from the PCF and the Greens, whose votes collapsed in 2002). Another reference point for comparison might be presidential and regional elections: in 2012, Hollande won 27.6% in the first round in the constituency; in 2010, the PS list had won 33.8% in the constituency for the first round of regional elections; in 2007, Royal won 25.8% in the first round.
As in the last by-elections, an important share of the PS’ 2012/2007 voters likely did not vote in the by-election. However, unlike in past by-elections, the first round results do not indicate a clear correlation between higher abstention and a higher share for the left either in the by-election or in 2012. In fact, turnout was generally lower where the FN did best. Furthermore, abstention in the first round (54%) was much lower here than in the Oise (67%) – sure, the Oise is more abstentionist by nature, but we might have expected that turnout would be rather low here considering circumstances: a large left-wing electorate (one much larger than in the Oise) tempted to not show up, a little-known PS candidate and the stench associated with the Cahuzac scandal.
We may reasonably assume that some ‘lost’ PS voters did in fact turn out, but voted for other leftist/left-leaning candidates: either the FG or EELV candidates but also Anne Carpentier, the aforementioned left-inclined satirical candidate who won 3.3% in the first round (more than the EELV candidate). There is some evidence, it would seem, that the FN’s strongest gains came from precincts where the left had done well in its happier days – the PS will hate to admit it, but some of its past voters voted for the FN this year.
As always, analysis of this election is clouded by partisanship, mindless blabbering by the inevitable self-proclaimed ‘experts’ and journalists who have no clue about elections coming up with their grand theories. One particular factor merits attention, to help explain the PS’ catastrophic result. The PS leadership has preferred to go the easy way out, finding scapegoats for its (largely self-induced) defeat: the Greens split the vote (reasonable, but the PS-FN gap was 770 votes and EELV won 914 votes, meaning that 84% of the EELV’s voters would have needed to vote PS for it to qualify) and it’s all Cahuzac’s fault. Cahuzac likely contributed to aggravating the situation for the PS, but the PS’ defeat was self-induced, by external (the government’s unpopularity, the crisis) and internal factors.
At the local level, the PS had a mediocre candidate who lacked an obvious geographic base in the constituency. Cahuzac had a local base in Villeneuve-sur-Lot, and he did a few points better in Villeneuve-sur-Lot than the constituency as a whole. Jean-Louis Costes, the UMP’s candidate, is mayor and general councillor of Fumel, a strongly left-wing town (62% Hollande). This hadn’t been of much help against Cahuzac in June 2012, although he did manage to hold Cahuzac down to 56% in Fumel (the commune). However, in a by-election against a little-known PS candidate with no similar local implantation, this was a major boon for Costes: he won 49.8% in the canton of Fumel, against 22% for the PS and 15.2% for the FN. In contrast, Barral (PS) peaked at only 28% in Tournon-d’Agenais and his support generally ranged from 20% to 25% in almost every other canton.
The UMP’s result, by itself, was nothing exceptional. In past by-elections, despite the unpopularity of UMP president Jean-François Copé since the November 2012 kerfuffle, the UMP’s vote had increased by a fairly substantial margin in the first round – on average by 7 or 8%. The only exception was the last by-election in the 8th foreign constituency, but that was most likely due to the right’s eternal internecine warfare there and very low turnout. Yet, in the Lot-et-Garonne, Costes won only 28.7% – only a marginal improvement on his 27% from June 2012, a result which had been very poor – in 2007, the UMP won 42% in the first round and in 2002, 30%. Some right-wing votes were probably dispersed between the many minor right-wing/centrist candidates and, to a certain extent, the FN.
The Left Front (FG) was, once again, unable to benefit from the PS’ collapse – its candidate won only 5.1%, a minor improvement (0.6%) on a fairly paltry showing in 2012. This harks back to two episodes where the PCF was in opposition to a PS government – from 1984 to 1986 and from 1988 to 1993. However, in both cases, the PCF was only able to stop the bleeding – it did not gain (a significant amount of) votes.
In the second round, Costes (UMP) won with 53.8%, a slightly wider margin than the UMP’s victory in Oise-2 (51.4%) and, on the whole, a fairly good result for the UMP given that the FN had a lot of momentum going into the runoff and many were wondering whether or not the FN would be able to pull off an upset victory.
Nevertheless, it is another excellent result for the FN – after gaining 10 points from 2012 in the first round, the FN increased its vote in the runoff by some 20 points (nearly 8,000 additional votes). What is more, unlike in the Oise-2, the FN candidate won more votes in the runoff (15,647) than Marine Le Pen had won herself in April 2012 (some 13,000). The UMP candidate gained 8,762 votes between the two rounds.
While left>FN transfers undeniably exist, the June 2012 legislative election showed that they were far less significant than right>FN transfers. In 9 right/FN runoffs in the last legislative election, there was only a weak correlation (0.21) between left-wing strength in the first round and FN gains between both rounds; there was, however, a 0.64 correlation between left-wing strength and a decline in voter turnout between between both rounds. Turnout declined by an average of 8% in the 9 right/FN battles in June, it only increased by 1.2% in left/FN battles. The percentage of voters who turned out in the runoff but cast blank or invalid votes was also very high (over 10%) in right/FN runoffs.
However, as in Oise-2, turnout increased (pretty significantly here, to the point where turnout was higher than abstention) between both rounds although there was, a large increase in invalid votes. Once again, this begs the question – where did the FN’s nearly 8,000 new voters come from?
In the Oise, an ecological inference analysis by Joël Gombin had found that 43% of the PS candidate’s first round voter had voted for the FN candidate in the runoff, with the remnants split between staying home, invalid votes and the UMP. However, a study done by the PS federation in the department disproved this ‘transfer’ theory in favour of the ‘substitution’ theory which holds that there was a significant change in the composition of the electorate between the two rounds, with first round leftists not voting being compensated by a large increase in turnout on the far-right. Analyzing 84% of the listes d’émargement (signing sheets where voters sign their initials after casting their ballot), the PS found that there was a large change in the electorate – about 4,700 first round voters did not vote a week later, but about 6,350 voters who had voted in the first round did so in the second round. This observation was more in line with the results of the 9 right/FN runoff in June 2012.
It is quite possible that the same thing happened in the Lot-et-Garonne, but there is also a strong possibility that a significant number of left-wing voters from the first round voted FN in the second round. If this was true, this would be a major defeat for the old strategy of the ‘republican front’ (anti-FN alliances). Unlike in the Oise, where the local PS candidate had not endorsed either the UMP or the FN candidate, the PS here endorsed the UMP candidate and Costes – fairly ironically given how he’s on his party’s right (MIL) – embraced the ‘republican front’. The ‘republican front’ strategy has been challenged and almost thoroughly discredited since 2010. On the one hand, the UMP no longer automatically endorses the left against the FN and many UMP leaders – Copé first and foremost – have had ambiguous statements on all this. The UMP nowadays tends to prefer the ni ni strategy – neither the left nor the FN – although the party remains split between a moderate faction of the ruling elite which still has sympathy for the ‘republican front’ and a more conservative activist base which has a large minority favouring open electoral alliances with the FN. The PS, meanwhile, still has a preference for the ‘republican front’ but the UMP’s strategy has unnerved it, to the point where some local PS candidates will endorse neither the UMP nor the FN. Recently, there were allegations that the PS in the Vaucluse covertly supported FN candidate (now deputy) Marion Maréchal-Le Pen by not withdrawing its candidate from the three-way runoff in which Marion Maréchal-Le Pen emerged victorious.
Finally, the continuation of a ‘republican front’ strategy tends to play right into the FN’s hand. A large part of Marine Le Pen’s rhetoric is denouncing the corrupt ‘UMPS’ elites – a message incessantly regurgitated by her new circle of obedient young leaders and candidates, including the FN candidate in this constituency. A ‘republican front’ between UMP and PS can easily be presented by the FN as ‘proof’ that both parties are, in reality, two sides of the same coin and are in cahoots with one another. And neither the UMP nor the PS try very hard to disprove that – PS deputies recently found common cause with UMP deputies in significantly watering down the government’s post-Cahuzac transparency and ethics legislation.
In the first round, the left (PS-FG-EELV) won around 10,400 votes (11,872 if you include Carpentier, the NPA and the Pirates). There were an additional 3,984 invalid votes in the second round, and we can safely assume that most of those were from left-wingers. 992 additional valid votes were cast in the second round. The UMP candidate gained 8,762 votes – the additional votes from the MoDem and the other right-wingers/centrists give him an additional 2,437 votes from the first round. There are therefore about 6,300 additional UMP votes which came from other ‘sources’. The FN candidate gained about 8,000 new votes. On these numbers, it would appear that the FN gained a large number of its additional votes from left-wing voters. But this is an extremely rudimentary and unscientific calculation. It can neither prove nor disprove the ‘transfer’ or ‘substitution’ theories.
As in the Oise-2, this by-election has shown two things – the PS is unpopular and faces an electoral drubbing if these numbers hold up in a national election; the FN is the only political force in the country which is truly on the upswing and it has proven that it has a remarkable ability to gain significant support from one round to another in duel runoffs. The cordon sanitaire is – in good part – gone. The FN has a far less ‘toxic’ image. Marine Le Pen’s dédiabolisation efforts are paying off, and many voters – left and right – are willing to vote for the FN over a more ‘acceptable’ party in the runoff when their preferred candidate is eliminated. We cannot treat voters as mathematical, rational and predictable individuals who can be expected to follow the directions given by their party of choice. Despite the strong enmity between national PS and FN leadership, there is some overlap between both parties. Some left-wing voters will prefer the FN over the right when faced with that choice.
The PS has lost four seats in by-elections, two of those were lost by the first round. PS candidates lost votes in all seven ‘normal’ by-elections, in all but one they lost a significant amount. The PS was eliminated by the first round in a total of four out of these seven by-elections. In France, midterm by-election loses for the governing party are the rule, so this is not particularly surprising although still quite spectacular.