In 2013, our political economy model correctly forecast Angela Merkel would lead a coalition with the SPD. Previously, from 1998 to 2009, we obtained successful forecasts running a single-equation vote function based on the incumbent coalition, except in 2002. There, the model had not been able to anticipate the narrow defeat of the FDP by the Greens. In short, this methodology was not accurate enough to capture both the mechanics of proportional representation and the influence of “hinge parties”. Therefore, in 2013, we opted for a SUR (Seemingly Unrelated Regression) model simultaneously explaining the vote for the main German political parties. In 2017, this reasoning is reinforced by the upsurge of AFD ^{Footnote 1} (Alternative Für Deutschland) and the “rooting” of Linke Partei in the German political life. Today, these two blocs are gathering almost 20% of the vote share in polls and could prevent the main parties (CDU/CSU and SPD) from reaching a majority. Thus, a correct forecast should take into account whether hinge and protest parties will be able to obtain 5% of the votes cast in September 2017. On this basis, a swing ratio turning the votes into seats should help us calculate all the feasible coalitions and infer who will be the next Chancellor.

From 1961 to 2013, the SUR Model ^{Footnote 2} is built upon five equations explaining the vote. In order to keep the logic of incumbency for the main coalition party, the INC-MAIN variable includes the vote for the party of the CDU or SPD Chancellor. Similarly, the OPP-MAIN variable contains the vote for the party in favor of the main opponent of the outgoing Chancellor. Afterward, specific vote functions for the FDP, the Greens (GRUNE), the Linke (ex-PDS) (LINKE), and the other parties among which AFD (OTHERS) are added.

The modeling of the Chancellor party and the opponent party obeys the government responsibility hypothesis. Then, a higher prior unemployment level (U_Q-2 and U*_Q-2 measured two quarters before the election ^{Footnote 3} ) should penalize the main incumbent party, while it should favor the main opponent. Likewise, the bigger the preference ^{Footnote 4} of the voters for the Chancellor (KANZ_Q-1 ^INC) or the opponent (KANZ_Q-1 ^OPP), the bigger the electoral premium. Furthermore, following the localization of economic responsibility (Powell and Whitten Reference Powell and Whitten1993), the hinge or protest parties should be less held responsible for the economic situation compared to the main ones. Therefore, the vote for the FDP should namely reflect the voters’ wish to have it in a future coalition. ^{Footnote 5} Consequently, we have distinguished between situations where the FDP was an outgoing party with the CDU (CO_INC ^FDP/CDU), or with the SPD (CO_INC ^FDP/SPD), or an opponent with the CDU (CO_OPP ^FDP/CDU).

Voting for the Greens is assumed to be mainly based upon their electoral potential, (PGRUNE) in the polls (IFD Allensbach and ZDF-Politbarometer), distinguishing whether they are opponents or incumbents. With regard to the Linke (PLINKE), we selected the voting intentions in the polls (IFD Allensbach and ZDF-Politbarometer) and a dummy variable coding the period before and after the Oskar Lafontaine takeover in 2005 (LAFONTAINE). Finally, the vote for the other parties (OTHERS) is largely explained by the vote intentions in the polls for the residual parties (POTHERS), especially the AFD (IFD Allensbach and ZDF-Politbarometer). Several times in German history, small residual parties approached scores between 2% and 5% thereby reducing the scores of the big parties. ^{Footnote 6} But overall, since 2013, one must notice the “nuisance power” exercised by the new anti-Euro German party AFD on CDU’s potential vote share. In order to estimate the predicted AFD vote share, we forecast first the OTHERS score. Thereafter, we calculate the average score for others given by German polls on average, excluding AFD. As a last step, we infer that AFD (forecast) = OTHERS (model forecast) – POTHERS (average polls, AFD excluded).

It should be noted that in the specification of the equations we have taken into account several political-institutional events (from the period 1961–2013) that affect the German elections. In this way the Grand coalition CDU/SPD (GCOAL²⁰⁰⁹) has had a negative impact on the CDU and even more on the SPD, but it has had a positive effect on the residual parties (OTHERS). Lastly, the first elections following German unification (REUNIF) in 1990 broke the Greens electoral progress, which has stayed constant until now.

After identifying these specifications for each of the vote equations, separately tested, we estimated the following SUR model. Then, from the estimated coefficients, we provide a forecast for vote shares and party seats for 2017.

The SUR model reads as follows (see table 1): In order to make the vote forecast for the parties ^{Footnote 7} , we have plugged in the following values : U_Q-2= 6.3 ; KANZ_Q-1 ^INC = 38% (A. Merkel); KANZ_Q-1 ^OPP = 49% (M. Schulz); CO_INC ^FDP/CDU= 7% ; PGRUNE^OPP= 7.6% ; PLINKE=8.6% ; POTHERS = 15.8% (wherein non-AFD = 5.2%).

Table 1

SUR Model for 2017 German Election

Note: The NOLINKE variable is scored 1 from 1961 to 1987 (zero otherwise) and it indicates that the LINKE (ex-PDS) is not present. The variable NOGRUNE is scored 1 from 1961 to 1976 (zero otherwise) and it indicates that the Greens are not present.

In order to make the vote forecast for the parties ^{Footnote 9} , we have plugged in the following values : U_Q-2= 6.3 ; KANZ_Q-1 ^INC = 38% (A. Merkel); KANZ_Q-1 ^OPP = 49% (M. Schulz); CO_INC ^FDP/CDU= 7% ; PGRUNE^OPP= 7.6% ; PLINKE=8.6% ; POTHERS = 15.8% (wherein non-AFD = 5.2%).

If the elections were held now (in March 2017), the transfer from the vote share into seats via the swing ratio makes clear that Angela Merkel would be reappointed narrowly as Chancellor at the head of a Grand coalition with the SPD (see table 2). ^{Footnote 8} Indeed, considering the AFD could gain 85 seats, a CDU-SPD combination seems to be the only feasible way to reach a majority (see table 3).

Table 2

Political Economy Model (SUR)

(*) Based on the 2013 seats share.

Table 3

Hypothetical Coalitions

(*) Based on the 2013 seats share.

Potentially, Angela Merkel could build a rightist majority (CDU-FDP-AFD) but AFD positions on Europe and immigration make it impossible. A jamaïcan coalition (CDU-Liberal-Democrats-Greens) would not also achieve an absolute majority.

On the other hand, a “red coalition” (SPD + GRÜNE+LINKE) would fail by 33 seats.

Thus, at the time of writing, the CDU-SPD is the only rational solution. However, Angela Merkel’s fate is not sealed once and for all. First, preference for Martin Schulz as the next Chancellor has clearly grown (40% in January and 49% in February). Our simulation shows a 60% rate would push the SPD ahead with 30.9% of the vote share. But this would not be enough for a plurality in seats. Secondly, Angela Merkel could be endangered if the FDP does not reach the 5% bar, in that almost 40 seats would be redistributed. Lastly, a deterioration in the unemployment rate (more than 1%) could further weaken Angela Merkel. Such an unfavorable aligning of planets would help Martin Schultz become the ninth postwar Bundeskanzler.