Provincial election results since 1993 for Bolzano-South Tyrol are now available in Elections to the South Tyrolean Landtag (Provincial Council). I have also included turnout statistics for the 2008 Landtag and Chamber of Deputies elections on The 2018 South Tyrolean provincial election, and updated the election statistics available there to fully account for the single-digit differences between preliminary and definitive 2018 vote totals.

Finally, I have corrected the correlations between voter turnout and percentage of Italian speakers at the municipal level in the 2013 and 2018 provincial elections. These were originally reported as -0.76 and -0.60, respectively, but they were actually stronger, standing at -0.84 and -0.70, respectively; the lower figures were due to the fact that five of South Tyrol's 116 municipalities had not been included in the original analysis. For comparison purposes, the correlation between voter turnout at the municipal level and percentage of Italian speakers in the 2008 Landtag election stood at -0.71 - almost identical to the corresponding figure for 2018.

Update: South Tyrol provincial election results since 1993 are now available on this website. This piece, originally posted on January 28, 2019, has been updated and corrected, as detailed on South Tyrol Landtag election results.

The province of Bolzano - South Tyrol, which held a provincial election last October 21, stands out from the rest of Italy in many ways. Besides being the country's northernmost province as well as its wealthiest one, South Tyrol is the only Italian province in which a majority of the population speaks the German language. Nevertheless, a substantial minority speaks Italian, and both languages enjoy official status along with Ladin, a Romance language related to Italian but distinct from it. In addition, the province houses a rapidly growing foreign population, which nowadays vastly outnumbers Ladin-speaking South Tyroleans.

Not surprisingly, the language divide has played a crucial role in South Tyrolean party politics, and the province has developed a multi-party system in which Italy's nationwide political forces - primarily supported by Italian speakers - compete with parties backed by German (and Ladin) speakers, most notably among them the South Tyrol People's Party (SVP), which has been the largest party in the province since 1948. In fact, SVP was a key actor in the establishment of devolved government at the provincial level in 1972; the region of Trentino-Alto Adige, formed by the provinces of Bolzano and Trento (Trent), had been granted devolution in 1948, but German-speaking South Tyroleans were not satisfied with this arrangement, primarily because they were (and remain) vastly outnumbered by Italian speakers at the regional level.

However, support for SVP has been gradually but steadily declining in recent years, due to the emergence of other parties within the German-language community, some of which advocate transforming South Tyrol into an independent country, or having the province reincorporated into Austria; the present-day region of Trentino-Alto Adige was part of the Austro-Hungarian Empire until 1919. Meanwhile, the environmentalist Greens have developed a not insubstantial following among members of all three linguistic groups. As a result, in 2013 SVP lost the overall majority it had held in South Tyrol's provincial council since 1948.

Although South Tyrol has often been cited as an example of peaceful coexistence between different ethno-linguistic groups, one notable political trend among Italian speakers - particularly in the provincial capital of Bolzano/Bozen, where they constitute a large majority of the population - was the unusually strong showing of the neo-fascist Italian Social Movement (MSI-DN) and its ostensibly post-fascist successor, the National Alliance (AN) between 1985 and 2008. MSI-DN, which remained a small yet not insignificant force in the rest of Italy, turned Bolzano - South Tyrol's largest city - into a party stronghold by exploiting ethnic resentment against the German-speaking population and its allegedly privileged position on account of the system of ethnic proportionality, which provides for the allocation of public goods (such as government jobs and welfare benefits) among the province's linguistic groups in proportion to their numerical strength. However, support for the hard right receded in the years following the National Alliance's participation in and subsequent merger with Silvio Berlusconi's right-of-center People of Freedom (PdL), and in the 2013 provincial election SVP narrowly topped the poll in Bolzano, five votes ahead of the Democratic Party (PD-DP), Italy's main center-left party.

That said, the last five years have brought about dramatic changes in Italy's party system. The emergence of the populist but ideologically ambiguous Five Star Movement (M5S) upset the dominance of the broad center-right and center-left cartels which had alternated in power since 1994, paving the way for a three-way race in 2013, in which the center-left coalition won a large majority in the Chamber of Deputies on a narrow popular vote lead, but fell short of a Senate majority. However, that state of affairs, as well as the subsequent, short-lived dominance of PD under the leadership of Matteo Renzi, turned out to be halfway houses. Much like France's Charles de Gaulle in 1969, Renzi bet his political future on an ill-advised constitutional reform, and just as in France nearly a half-century earlier, the gamble backfired spectacularly when voters rejected the proposal in a 2016 referendum. As a result, Renzi had no choice but to resign as head of government, dealing a heavy blow to the center-left in general and PD in particular.

By the time a general election was held in March 2018 under a new electoral system, the alliance formed by the center-left parties - still struggling to recover from the aftermath of the 2016 referendum - finished third, well behind the center-right coalition and M5S, which became Italy's largest single party. However, the center-right coalition fell short of an overall parliamentary majority in both houses of Parliament, and eventually a coalition government was formed by M5S and the Northern League, which had become the largest party within the center-right alliance, displacing Berlusconi's Forza Italia.

Meanwhile, South Tyrol remained one of the few areas in Italy where the center-left coalition prevailed, by virtue of an alliance with SVP. The latter, while not a leftist party by any stretch of the imagination, has usually aligned itself with Italy's center-left alliances. This is due in no small measure to the historically strained relationship between South Tyrol's German-speaking population and the Italian right, going all the way back to the Fascist dictatorship of Benito Mussolini: it is no exaggeration to say that to this day, some people in Italy show little if any contrition about the Fascist regime's attempt to forcibly Italianize South Tyrol, which entailed among other things banning the use of the German language in the public sphere, and the large-scale resettlement of Italians from other parts of the country, in order to dilute the numerical superiority of German speakers in the province.

Since the March 2018 general election, opinion polls in Italy have indicated that support for the right-wing Northern League has soared to the point it might become the largest single party should a general election be held in the immediate future. The League's rising electoral fortunes have been largely confirmed by recent regional, provincial and local elections, among them the provincial vote in South Tyrol last October. However, due to the province's unique political environment, the extent of this turn of events is not fully evident from the overall results.

As in previous elections, the official South Tyrol 2018 provincial election website provides aggregated results for the eight municipalities with a majority of Ladin speakers. However, I took that one step further, and aggregated the results for the five municipalities where according to the 2011 census a majority of the population speaks Italian, and for the 103 municipalities where German speakers are in the majority. The aggregated results for the three groups and for the entire province of South Tyrol (the latter including postal ballots), were as follows:

Now, it should be noted that these aggregations have the following constraints:

1. While the Ladin valleys are fairly homogeneous in terms of language, the Italian- and German-speaking municipalities have both significant linguistic minorities, particularly in the case of the former. According to the 1991 census language figures for South Tyrolean municipalities - the last census for which South Tyrol's Provincial Statistics Institute has municipality-level absolute language figures available online (for 2001 and 2011 language statistics at that level are available as percentages only) - the distribution of languages stood as follows:

   Language       Italian- speaking Municipalities       German- speaking Municipalities       Ladin- speaking Municipalities       Valid Declarations                                           Abs.       %       Abs.       %       Abs.       %       Abs.       %       Italian       80,300       71.66       36,093       12.28       521       3.09       116,914       27.65       German       30,920       27.59       255,691       86.99       892       5.28       287,503       67.99       Ladin       834       0.74       2,133       0.73       15,467       91.63       18,434       4.36

   Note that the language distribution of South Tyrol's non-foreign population hasn't changed significantly since 1991. According to the 2011 census, the province had 118,120 Italian speakers (26.06%); 314,604 German speakers (69.41%); and 20,548 Ladin speakers (4.53%), for a total of 453,272 valid language declarations. Moreover, available 2011 figures for the Ladin valleys indicate the area remained 90.85% Ladin-speaking at that point in time. It should also be noted that the cited language figures include a small percentage of aggregation declarations made by South Tyroleans who don't consider themselves as belonging to any of the three official linguistic groups, but who are nonetheless required by law to specify the group to which they wish to be aggregated.

2. In 2011, the percentage of German speakers in the group of German-speaking municipalities ranged from 50.47% in Merano/Meran - South Tyrol's second largest city, where 49.06% of the population spoke Italian - to 100% in Martello/Martell, the only municipality in the province where the entire population spoke a single language; leaving Meran out of this category, the remaining municipalities were 91.28% German-speaking and 7.97% Italian-speaking in 1991.

All the same, the figures show a dramatic contrast between the Italian-speaking municipalities on the one hand, and those with German- and Ladin-speaking majorities on the other one. In the former, the League topped the poll with nearly twice as many votes as SVP, and over thrice as many as PD. In fact, the results were broadly similar to those of the provincial election in neighboring Trento province, which also went to the polls last October 21; although recognized linguistic minorities in the latter constitute just four percent of the population, the province has a significant autonomist party, namely PATT, which is also a frequent ally of SVP at the regional level.

Meanwhile, in the German- and Ladin-speaking municipalities SVP remained by far the largest party, although the new Team Köllensperger (TK) list headed by Paul Köllensperger, a former M5S member, had a strong showing on its electoral debut and arrived second both in those areas and in the entire province. The League emerged as the largest Italian party in both the German and Ladin areas, albeit reduced to single digits and placing a distant fifth in the German-speaking municipalities. That said, no Italian list had managed to poll as much as five percent of the vote on that part of South Tyrol in the provincial elections held between 1998 and 2013. Just as important, the League finished third in the Ladin municipalities with 10.60% - a noteworthy outcome considering that Italian speakers constitute just 4.70% of the population in the Ladin valleys. The remaining Italian lists fared poorly on both German- and Ladin-speaking areas, but the Greens continued to draw support from all three linguistic groups, although their share of the vote was noticeably higher in the Italian-speaking municipalities.

One particularly divisive issue in the recently held provincial election was a proposal by Austria's far-right Freedom Party (FPÖ) - currently the junior partner in that country's right-wing coalition government - to grant Austrian passports to German- and Ladin-speaking South Tyroleans, but not to Italian speakers. The proposal has met strong opposition from the Italian government, but was embraced by South Tyrol's separatist parties. However, these parties - particularly Die Freiheitlichen (whose name is sometimes translated to English as "The Libertarians") - incurred in substantial losses. At the municipal level, these losses show a relatively strong inverse correlation (-0.64) with the results obtained by TK, particularly in the 103 German-speaking municipalities (-0.74).

The following table shows vote and percentage gains or losses for selected lists with respect to the 2013 provincial election (with postal ballots taken into account for the provincial totals):

Interestingly enough, in percentage terms SVP incurred a greater loss in the Italian-speaking municipalities, even though the party is far weaker in them than anywhere else in the province. In fact, it is the exact opposite of the pattern observed between the 2008 and 2013 provincial elections, in which the party's losses came primarily from the German-speaking municipalities. Just as important, SVP no longer commands an absolute majority in that part of the province, where as recently as 1998 its share of the vote stood at 70.18%; this is true as well even when leaving out bilingual Merano/Meran (in which case SVP's share of the vote in the remaining German-speaking municipalities increases to 49.34%, down from 73.94% in 1998). Meanwhile, the League's vote total had a larger growth rate in the German-speaking areas, even though its vote percentage increase was by far bigger in the Italian-speaking municipalities.

While voting patterns in South Tyrol largely follow the province's linguistic distribution, the relationship is not strictly one-to-one. There have been instances in which a majority or nearly a majority of voters on some Italian-speaking municipalities have supported ethnic German lists in provincial elections, while in national elections to the Italian Chamber of Deputies, the combined share of the vote for Italian party lists in the German-speaking municipalities has usually exceeded the reported proportion of Italian speakers in that part of the province by a significant margin. In fact, in some German-speaking municipalities with a substantial Italian-speaking minority, Italian parties have often outpolled ethnic German parties in Chamber elections.

The described electoral behavior could be explained by a number of factors, such as population shifts in the years since the 2011 census; linguistic distribution disparities between the overall and voting-age populations; individuals who indicate a linguistic group affiliation different from their first language (or languages); voters supporting a party outside their linguistic group (or an inter-ethnic list such as the Greens); and voter turnout disparities between linguistic groups. The first three factors are beyond the scope of this posting, but official election statistics confirm the importance of the remaining two.

From 1948 to 1998, voter turnout in provincial as well as national elections in South Tyrol stood at exceptionally high levels, usually above ninety percent. A slight decline in turnout became evident by the end of the century, but as late as 1998 turnout differences between South Tyrol's eight districts remained small, standing at just over five percent. Even so, in the provincial election held that year voter turnout in the district of Bolzano - coextensive with the provincial capital - fell at a noticeably higher rate than in the rest of the province, a trend which rapidly accelerated in subsequent provincial elections. By 2013 the turnout gap between districts exceeded twenty percent, and turnout in the Italian-speaking municipalities was almost sixteen points below the corresponding figure for German-speaking municipalities. Meanwhile, voter turnout rates for Chamber of Deputies elections held during that period showed very little variation between Italian- and German-speaking municipalities, with differences of less than one percent between 2006 and 2013.

However, the 2018 legislative and provincial elections were both characterized by a drastic departure from previously observed patterns. In the Chamber of Deputies vote last March, voter turnout in South Tyrol fell sharply with respect to the preceding 2013 election, but the drop was particularly acute in the German- and Ladin-speaking municipalities; as a result, voter turnout in Italian-speaking municipalities was significantly higher than in German-speaking municipalities; the latter also registered a large increase in the number of invalid ballots. This unusual turn of events appeared to be due to the non-participation of South Tyrol's separatist parties. But in the October provincial election turnout rose slightly in the Italian-speaking municipalities, relative to the 2013 provincial vote, while it fell significantly in German- and Ladin-speaking municipalities. As a result, the turnout gap between districts shrunk to just over fifteen percent, while the difference in turnout between Italian- and German-speaking municipalities came down to less than twelve percent. All the same, voter turnout in South Tyrol has now declined nearly twenty points over the course of four decades, from a record high of 93.21% in 1978 to 73.87% this year (excluding postal ballots). For the Landtag (Provincial Council) and Chamber of Deputies elections of 2008 to 2018, voter turnout in South Tyrol and the three municipality groups stood as follows:

In the 2013 provincial election there was a fairly high inverse correlation (-0.84) between voter turnout and percentage of Italian speakers at the municipal level. However, by 2018 the correlation had dropped to -0.70, and there was a moderate correlation (0.58) between change in voter turnout and percentage of Italian speakers. At the municipal level, eighteen of the top twenty voter turnout rate decreases (all above seven percent) took place in municipalities which were at least 95% German-speaking, while at the district level, Val Venosta/Vinschgau (97.29% German-speaking) had the largest drop (-7.10%) in turnout. Meanwhile, the turnout rate in Bolzano remained unchanged to two decimal places. These figures suggest that while turnout increased among Italian speakers, it went down among German speakers, quite likely outpacing the rising numbers among the former. As a result, in the Italian-speaking municipalities the increased turnout among Italian speakers was largely but not completely offset by the lower turnout among German speakers, resulting in a minimal overall increase. However, in the German-speaking municipalities the turnout increase among the proportionally much smaller Italian-speaking minority barely dented the falling turnout among German speakers. In terms of Landtag election results, the shifting turnout trends may explain why ethnic German parties outpolled Italian lists in the three smaller Italian-speaking municipalities in 2013, but not in 2018.

Voter turnout also fell noticeably - by an average of more than six percentage points - in all but one of the South Tyrolean municipalities bordering Austria (all of them overwhelmingly German-speaking). The one notable exception to this trend was Brennero/Brenner, named after the Brenner Pass, an Alpine mountain pass and a major transportation link with Austria; SVP substantially increased its share of the vote in the municipality on a turnout rate slightly larger than in 2013.

While the Austrian passports controversy may have played a role in the latest decline in voter turnout, the relationship between both developments is not quite clear. To be certain, the percentage decline of Die Freiheitlichen, when calculated on the basis of the entire electorate (that is, including non-voters and invalid votes), shows a mild correlation (0.42) with the decrease in turnout, but the correlation is strongest (0.84) in the eight municipalities where at least one-third of the population is Italian-speaking.

Although the results polled by the League in 2018 broadly resemble those obtained in 1998 and 2003 by the National Alliance - the last significant Italian hard-right party in South Tyrol until the League's recent breakthrough - the distribution of votes in the three linguistic-based municipality groups, presented below, show a number of differences.

As previously noted, the League's share of the vote in the Ladin Valleys stood substantially above the percentage of Italian speakers in the area, which suggests that a number of Ladin-speaking voters backed the party as well. This appears to have been the case as well in the municipality of Castelrotto/Kastelruth, which is predominantly German-speaking but also has a significant (15.37%) Ladin minority. In fact, the League vote in that municipality is concentrated on its sixth polling section, which in past provincial elections has registered substantially above-average support for Ladin lists. However, support for the National Alliance in South Tyrol's predominantly Ladin-speaking areas was negligible in 1998 and 2003.

Meanwhile, the League's vote in the predominantly German-speaking areas of the province stood slightly above the figures achieved by AN in 1998 and 2003, but the League's share of the vote also exceeded the percentage of Italian speakers in forty-one German-speaking municipalities without a significant Ladin-speaking population. In these municipalities - 1.79% Italian-speaking in 1991 - AN nosedived to just 0.42% and 0.36% in 1998 and 2003, compared to 2.88% secured by the League in 2018. At the district level, the same trend is particularly evident in Val Venosta/Vinschgau, where the League won 2.62% of the vote in 2018, even though only 2.63% of its population was Italian-speaking in 2011 (down from 3.06% in 2001 and 3.41% in 1991). By comparison, AN polled just 0.86% of the district vote in 1998, and 0.67% in 2003. Thus, it would appear that a few German-speaking voters - or at least voters who identify as belonging to the German-speaking group - also backed the League. In fact, in the municipality of Martello/Martell, whose entire population was German-speaking in 2011, the League won 2.14% of the vote. However, the League represents a different brand of hard-right politics, inasmuch as in the recent past it had been a separatist party advocating the independence of northern Italy from the rest of the country, which put it at odds with the National Alliance. Moreover, in the 2014 European Parliament election the League ran in coalition with Die Freiheitlichen.

As a result of the 2018 provincial election, the distribution of seats in the South Tyrolean Landtag, allocated on a provincial basis by the largest remainder method of proportional representation, stood as follows:

Compared to the 2013 provincial election, SVP lost two seats while PD-DP lost one, and the two parties no longer commanded a Landtag majority. Since the government of South Tyrol must reflect the province's ethno-linguistic composition, SVP had only two choices: either form a coalition cabinet with the League, or include the Greens in the existing coalition with the Democrats. In the end, SVP leaders concluded the former was the least problematic of the two options, and opened negotiations with the League in the weeks following the election, which concluded in the formation of the province's first-ever SVP-League government on January 25, 2019.

However, the lengthy negotiations between SVP and the League were nearly derailed last December by the Italian government's constitutional reform proposal to decrease by one-third the size of both houses of the Italian Parliament. The original proposal called for the number of Senate single-member seats in South Tyrol to be reduced from three to two, but this ran counter to a 1991 law which fixed at three the number of such seats for each of Trentino-Alto Adige's two provinces. Because said law was part of a package of measures favorable to the population of Alto Adige, which led to the settlement of the South Tyrol issue with Austria in 1992, the proposed constitutional reform threatened to re-open a very sensitive issue, not least because under the existing arrangement, two of South Tyrol's three Senate seats are overwhelmingly German-speaking, while one has an Italian-speaking majority, which approximates the province's linguistic makeup. In the end, a compromise was reached whereby autonomous provinces would be guaranteed a minimum of three single-member seats, thus restoring compliance with the existing provisions. Even so, the controversy underscored the fragility of the coalition agreement between SVP and the League. Time will tell if this arrangement will prove to be stable or long-lasting.

Interestingly enough, one or other of the two independent candidates emerged as the second choice of all voters who indicated a valid second preference, irrespective of their first preference choice. A plurality of Poliquin and Bond first preference voters chose Hoar, while a majority of voters selecting Golden or Hoar as their first preference went for Bond, who obtained the largest overall second preference total. Nevertheless, it must be emphasized that the second preferences of Poliquin and Golden voters had no bearing on the election outcome, which was determined by the elimination of independent candidates Bond and Hoar after the first preference count, and the addition of their second preferences to the first preference votes cast for continuing candidates Golden and Poliquin, including 6,615 of the 8,260 ballots with first and second preferences for the two independents plus valid third preferences for Golden or Poliquin.

An overwhelming majority of voters indicating valid preferences for at least two different candidates - 136,895 out of 140,325, or 97.6% - cast valid first and second rankings. Of the remaining 3,430 ballots, a total of 2,604 had valid first rankings, an undervote (that is, blank) second ranking, and a valid third ranking for another candidate. Following Maine's RCV counting rules, single skipped rankings were ignored, and the third ranking was counted as the second preference; the same logic was applied to 259 ballots with an undervote on the first ranking, but valid preferences for different candidates in the second and third rankings (which were counted as first and second preferences).

The statistics presented here were obtained from CSV-format data available in State of Maine 2018 Ranked Choice Voting (RCV) Election Data, which has additional information and links pertaining to the election.