Crunching numbers (preliminary thoughts)

04.16.2004

I've started crunching electoral statistics. Still only early on, since I first have to manually input all the electoral data into various Excel spreadsheets. That's six elections (1980, 1985, 1989, 1993, 1997, 2002), nine national departments, and two elections (1997, 2002) after Bolivia adopted the MMP (Multi-Member Proportional) electoral system similar to Germany's (creating 68 single-member districts). Not done crunching the numbers for various statistics (especially at the disaggregated level), but some preliminary observations:

The idea for adopting the MMP system — which uses a two-tiered ballot structure — was to increase the connection between parties & voters. In MMP systems, voters elect a district representative; the second part of the ballot completes the lower house w/ proportional representation (PR) to compensate. Essentially, MMP is a variation on PR. It's supposed to make parties work harder at establishing (& keeping) local constituencies.

Seems the plan backfired. Introduction of the MMP system increased the effective number of parties (increasing political fragmentation) — while discouraging large parties from heavily engaging in local districts (since larger parties only have to win large chunks of PR-list votes) and encouraging small parties (since winning a "uninominal" (single-member district) seat can take as little as 18%). The new rules essentially reward small, radical, clientelistic parties.

Even more ironic, blank votes (which are high & increasing after adopting MMP) are actually higher in the uninominal part of the ballot. I've not yet worked the figures for Potosí (the department w/ highest blank ballots), but in 1997 Chuquisaca, one uninominal contest reported 13.3% blank votes. Adding null votes, it's 17.1%. That means 1 in 6 voters didn't vote for any of the 12 potential candidates in that district. Insane.

I'll join the ILDIS researchers who recently decided (for different reasons) to suggest abandoning MMP & returning to a more traditional PR electoral formula.

Posted by Miguel at 04:59 PM

Comments

Sme questions abt your dissertation:
Those public intellectuals that u've been interviewing, are they the "technocrats"? Why are their views significant in your field work? How important is their role in influencing public policies? Any instances?

Posted by: Stephanie at April 18, 2004 11:02 PM

Well, some of these public intellectuals are "technocrats", but they do have a powerful influence. President Mesa was a public intellectual before becoming the vice presidential candidate. Many leaders of political parties (especially MBL) are public intellectuals. The conference I went to on decentralization was significant; it was intended to give the government a recommendation (the entire event was taped, transcribed, and sent out).

About 20 years ago there was a conference on reforming the electoral system. Participants included Gonzalo Sanchez de Lozada (later president, twice), Carlos Mesa (now president), Ronald McLean Abaroa (later presidential candidate, mayor of La Paz), Filemon Escobar (later senator for MAS), Victor Hugo Cardenas (later vice president). The topic of the conference: Whether or not to adopt the German MMP electoral system. The system was adopted by the Goni-Cardenas administration.

Posted by: Miguel at April 19, 2004 03:20 PM

It's certainly gratifying when some suggestions of these public intellectuals are implemented or at least taken into account.

I thought electoral systems were set in stone. What was the electoral system before MMP?

Probably a silly question: what are blank votes and null votes?

Posted by: Stephanie at April 20, 2004 07:40 AM

No, electoral systems aren't set in stone, and are often modified in many countries. The Bolivian electoral system's gone through many modifications. Prior to being an MMP system, it was a PR (proportional representation) system, though it changed from a variation on lowest cuotient remainder to Sainte-Laguë to d'Hont (the three are different counting rules used in PR), before changing to MMP.

Blank votes are votes cast w/ no marks on them, which are often protest votes. In countries where voting is compulsory, voter turnout is high, but you also have a good number of blank votes. Null votes are sometimes also called "spoiled" ballots. These are votes that are improperly marked (like marking two candidates, instead of one). The final tally count for seats is counted over "valid" votes — all votes properly cast (neither blank nor null).

Posted by: Miguel at April 20, 2004 05:20 PM

"Many leaders of political parties (especially MBL) are public intellectuals"

Interesting... the government is made up of many academics. Why then is Boliva in a dismal state?

(Don't mean that question to come across as sarcastic and/or simplistic - just straight to the point)

Just puzzled abt the economic and social problems besieging Bolivia, when there's a thinktank of intellectuals and their recommendations are frequently taken up.

Posted by: Stephanie at April 22, 2004 12:55 AM

Just to let you know, since u post often abt the neighboring conflicts between Chile and Bolivia.

I received an invitation (from the Institute of SEA Studies) to a lecture next Friday by the president of Chile, Ricardo Lagos. The topic is on global challenges of the 21st century (THE hot topic for the next 96 years).

I was reading the info about his background in the pamphlet. He too was an academic before he got into politics (which might explain my earlier comment).

Posted by: Stephanie at April 22, 2004 01:16 AM

Well, to answer your question most directly: One can be an intellectual w/o being intelligent.

Or, more precisely, politics & academia aren't the same thing. Most Bolivian "public intellectuals" aren't really "academics" in the American sense. Most political science in Bolivia is really just punditry.

The old joke that the only numbers are the page numbers. I've only seen one work that even used regression analysis — and it didn't provide any informatino on the kind of regression used (bivariate? multivariate? two-tailed?) or provide p values (which tell you if there results are statistically significant. Very poor quality; the results shouldn't be trusted.

But many "public intellectuals" are idealists who don't know how to play hardball politics. Read too much Marx or Rawls or French philosophers — not enough Machiavelli. That's Mesa's problem, for example. He seems to think he can solve the country's problems w/ a good, elegant, studied speech — not w/ political horse-trading.

Posted by: Miguel at April 22, 2004 12:18 PM

Interesting. I'd always thought lots of political problems arise because politicians are too Machiavellian.

Posted by: Stephanie at April 24, 2004 11:37 AM

Well, I think too many problems in politics are from not reading Machiavelli, not the other way around. At least from a practical standpoint. Many people don't like Machiavelli because he's not an idealist — he doesn't care to describe how politics should be, but how it is.

For example, one of Machiavelli's dictum's is "keep your friends close, and your enemies closer." Mesa should've known that. There were rumors (only rumors, mind you, no substantiation) that the viceminister of defense (a navy admiral) was involved in plotting a coup. Mesa removed him. Dumb move. If the guy was plotting a coup, better to keep him close & monitor him. If he wasn't plotting a coup, he now might be (upset at losing his job, public disgrace, etc.). See?

Posted by: Miguel at April 24, 2004 12:48 PM