On Sunday voters in El Salvador will elect their next president. As discussed here previously, I still think the leftist candidate, Mauricio Funes, will win, ending 20 years of rule by the right-wing ARENA party and marking the first presidential victory by the former guerrilla FMLN. But it could be close, and one can’t ever rule out a surprise.
One thing we know for sure is that there will not be a runoff. El Salvador’s presidency is elected by majority runoff and this is the first round. It is the first round of the presidential elections, that is. There were legislative elections in January, the first time that Salvadoran legislative and (first-round) presidential elections have occurred in the same year but not on the same day.
The reason there surely will not be a runoff is that the candidates of the other parties withdrew from the race after the legislative election. So, in essence, the legislative election functioned as the de-facto first round of the presidential election. The elections for legislature confirmed what surely everyone knew: that the FMLN and ARENA were by far the two dominant parties. More importantly, they perhaps showed the small left party, the CD, that it had better get out of the way and the vaguely centrist but mostly right-wing PCN and PDC that it might be possible to defeat Funes, after all. While a majority of votes would be required, whether in one round or two, the “momentum” factor of a strong plurality by either major candidate might have been something the other bloc wanted to avoid. Better to go for the one-round victory, when there is no question about who will contend for the second slot in the runoff (as there often is in two-round contests).
In any case, a legislative election within a few months of a presidential election is expected to function as a field-winnower. At least that is what I have long claimed about these “counterhoneymoon” elections. Nice to see theory confirmed by data–or, in this case, datum!*
______
See elsewhere:
-
POLL NUMBERS!!! Salvador election close, or not (boz)
Democracy Promotion in El Salvador: Elections 2009 (Latin American Musings)
El Sal’s street-level politics (The Democratic Piece)
* Actually, it is data, plural: two blocs of parties, and each winnowed its field to one presidential candidate.
Interesting. And I thought Australia led the world record for unnecessary duplication of election dates (eg, Queensland State election, 7 February 2004; Queensland local council elections, 27 March 2004).
Although plurality nationwide elections for Presidents are out of fashion (Iceland, and, uh…), holding a single-tick ballot for President shortly after a PR election for the Legislature would mitigate some (not all) of the lottery involved in first-past-the-post voting. I recollect reading that this was a factor in pre-1973 Chile – although (a) with 6-year presidential terms, there was a lot of time for voter preferences to shift, and (b) the system was not technically pure plurality – but popular vote phase was, and the Congressional “runoff” joint sitting, by convention, always voted to rubber-stamp the popular vote plurality candidate even if he was below 50%.
Seed planted by Tom Round — 14 March 2009 @ 20:46
The problem with the pre-1973 Chilean model was that it could produce a president on just over 1/3 of the popular vote. Arguably that was not always enough to conduct an effective government. Chile had a record of failed presidencies under those conditions.
Seed planted by Alan — 14 March 2009 @ 21:00
As Tom noted, Chile’s president under the 1925 constitution was elected by majority, not plurality. If no candidate won a majority, the decision passed to congress. It was by no means a rubber-stamp, especially in Allende’s case.
These two cases (Chile from 1930s to 1970s and El Salvador since 1990s) are somewhat similar in the ideological breadth of the party system and the use of quite different term lengths for president and assembly.
Chile never had a legislative election this closely precede a presidential election, however. From 1937 to 1973 legislative elections were always in March and presidential elections in September. Never in the same year.
Of the 7 presidential elections in that period, 2 were won by absolute majority, one by a 40-30 margin, and another by 47-28. Only in 1958 and 1970 was the popular plurality under 40% or the margin within a few percentage points. So the system they had worked pretty well till 1970. That was a big failure, to be sure, and I have written about the institutional sources of failure both in academic publications and here at F&V.
Seed planted by MSS — 15 March 2009 @ 17:40
It does occur to me that the logic of counterhoneymoon elections that I articulated above (i.e. winnowing the field of presidential contenders) probably makes more sense with plurality rule for the presidency than with majority runoff (the Salvadoran case notwithstanding).
The only country that has used a counterhoneymoon cycle almost exclusively for a long period is Colombia. There the presidential election method was changed from plurality to majority runoff effective with the 1994 election. But we have continued to see some “primary” effects of the cycle (now overlaid with actual presidential primaries concurrent with the legislative election in some parties some years).
By the way, Alan, plurality elections for president are not really an endangered species, albeit they are a declining percentage of presidential systems. A few examples include some fairly large countries: Mexico, Philippines, S. Korea, Venezuela, Zambia.
Other than Iceland (where the presidency is hardly important) the only semi-presidential cases that uses plurality that I can think of are Taiwan and the Palestinian Territories.
(And I thought one of you would quibble over the hyphenless [hyphen-less?] ‘counterhoneymoon’.)
Seed planted by MSS — 15 March 2009 @ 21:04
Chile’s problems were compounded by the fact that Senate elections were staggered as well, with approximately half the members chosen every four years for eight-year terms.
For example, in the 1965 legislative election the Christian Democratic Party (PDC) – which had captured the presidency the preceding year – won an absolute majority in the Chamber of Deputies but ended up with less than a third of the Senate seats: in a nutshell, PDC had been a minor (if rapidly growing) party prior to its 1964-65 election victories, and had won few upper house seats in the preceding legislative election, held in 1961.
The lack of a Senate majority greatly complicated matters for the Christian Democratic government of President Eduardo Frei Montalva, as the opposition left- and right-wing parties would often join forces to block Frei’s reformist agenda in the upper house, albeit for diametrically opposite reasons: the left opposed Christian Democratic reforms as limited and insufficient, while the right was opposed to them because they upset the status quo. Interestingly, the Communist Party of Chile was more willing to find common ground with the Christian Democrats on some issues (such as land reform) than Allende’s Socialist Party, which was more dogmatic than the Communists, and consequently downright obstructionist.
Incidentally, a similar problem would have probably dogged Allende had a legislative election been held in 1971 with the same results as that year’s municipal vote.
Seed planted by Manuel Alvarez-Rivera — 15 March 2009 @ 23:14
Oddly enough the staggered terms in the Australian senate have almost the same effect as in Chile before 1973. The Rudd government’s Carbon Pollution Reduction Scheme faces almost certain senate rejection because Greens think it too cautious and the Coalition think it too radical.
Seed planted by Alan — 16 March 2009 @ 02:29
My culpa (not Alan’s) re plurality presidents being an endangered species… although they certainly are declining. (Liberia too?)
Seed planted by Tom Round — 16 March 2009 @ 06:27
Liberia uses majority runoff. But I did add Taiwan to the (short) list, above, of semi-presidential systems with plurality election of the head of state.
Seed planted by MSS — 16 March 2009 @ 15:42
Almost the type case of the Shugartian subhyphenated counterhoneymoon,
Seed planted by Alan — 20 March 2009 @ 09:33