Incumbency Has Its Benefits
The article below is reprinted with the kind permission of the Hill Times, from their Monday April 27 edition.
By ALICE FUNKE and W. T. STANBURY
Recent news stories about conflicts within a few Conservative ridings over the party’s new incumbency protection policy, and about new preconditions being imposed on Liberal incumbents to have their nominations protected, once again implicitly raise the question of the electoral benefits of incumbency (see Abbas Rana, The Hill Times, March 16, 2009; Don Martin, National Post, March 27, 2009; Trevor Howell, Calgary Herald, March 29, 2009; and David Akin, Blog, April 6, 2009). The purpose of this piece is to provide data on the odds of incumbent MPs being re-elected versus the odds of challengers being elected, based on the 13 federal general elections between 1968 and 2008. Comparisons are made to the U.S.
Political party policies for incumbent MPs and nomination contests
In The Hill Times of March 16, 2009, Abbas Rana provided a convenient summary of the current policies of three major parties concerning nomination contests for incumbent MPs
New Democratic Party: “has a policy of not protecting their incumbent MPs from the nomination process and requires MPs to win open nomination contests prior to every federal election.”
Conservatives: “Riding association members can ask us for a race and if twothirds or more of the membership in any riding asks for a nomination race then there will be a contest.” (Riding members receive a ballot in the mail to decide whether to hold a nomination contest.)
Liberals: Incumbents must show to the central party headquarters by June [2009] that they have 400 members in their riding associations and at least 40 “Victory Fund” donors who donate $10 or more on a monthly basis, $5 to the party headquarters, and the other $5 to the riding association.
Bloc Québécois: According to La Presse, the Bloc started its own process of 75 nomination meetings last week, and expects to have 40/75 candidates nominated by September (Joël-Denis Bellavance, April 20, 2009)
Party policies have varied over time. Andrew Steele (blogpost in The Globe and Mail, April 7, 2009) pointed out that “In 2006, the [Conservative Party] did not allow any contest against sitting MPs. In 2008, the window for contesting nominations was peculiarly short.”
Further, “the Liberals automatically renominated all incumbents in the 2000 election, citing electoral urgency because the Prime Minister wanted to call an election.”
John Ivison (National Post, March 10, 2009) notes that “at the Liberal National Convention… in 1993, Jean Chrétien rammed through a resolution that deprived grass-roots Grits of the only power they had—the ability to choose their candidate during a nomination process…. Mr. Chrétien said he would protect the nomination of MPs indefinitely, provided they did not speak out against him.”
Re-election rates for incumbent MPs between 1968 to 2008
To examine incumbency rates (the percentage of incumbent MPs seeking reelection who were re-elected) we gathered the data on 13 general elections between 1968 and 2008 (Table 1). The 13 elections resulted in five different types of government in Ottawa: six Liberal majorities; two Liberal minorities; two Progressive Conservative majorities (under Brian Mulroney); one PC minority and two Conservative Party minorities (under Stephen Harper). Thus five of 13 elections resulted in a minority government over the 40-year period with three since 2004.
The overall rate of re-election for incumbent MPs was 78.1 per cent. In only two of 13 elections (1984 = 60.2 per cent; 1993 = 41.1 per cent) did the incumbency rate fall below 70 per cent. For seven elections, the rate was over 80 per cent, and in 2000 it was just over 90 per cent (Table 1).
The low incumbency rate in 1993 (41.4 per cent) reflected a most unusual election. This first of three majority Liberal governments under Jean Chrétien saw the Liberals gain 177 seats from 83 in 1988. More importantly, two new parties had a big breakthrough. The recently-created Bloc Québécois, running a full slate of 75 candidates in Quebec won 54 seats. Thus a separatist party became the official opposition.
The Reform Party, created in 1987, went from zero seats in 1988 to 52 in 1993. The NDP collapsed from a record 43 seats in 1988 to only nine in 1993. However, the Progressive Conservatives under Kim Campbell (who had replaced Brian Mulroney) were almost wiped out, falling from 169 seats in 1988 to two in 1993. Thus five parties obtained seats in the Commons in 1993 (Table 1). That election also saw the lowest percent of incumbents seeking reelection in the period 1988 to 2008 (220 of 295 = 74.6 per cent), as compared with the high watermark of 2006 (278 of 308 = 90.3 per cent).
Evidence that 1993 was an anomaly can be found in the next election (1997) when the Liberals won only 155 seats (a drop of 22) in an enlarged House of 301 seats, but 82.6 per cent of Liberal incumbents were re-elected.
The other year in which the incumbency rate was low (60.2 per cent in 1984) also reflected highly unusual conditions. Brian Mulroney won the greatest sweep in Canada’s political history. The PCs won 211 of 282 seats while the Liberals won only 40. In the earlier 1980 election, the PCs had won only 103 seats compared to the Liberals 147. This number of first-time MPs in 1984 (133 of 282 = 47.2 per cent) was the second highest in the 13 elections reported in Table 1. The highest was in 1993: 199 of 295 or 67.5 per cent.
The high incumbency rate (an average of 78.1 per cent between 1968 and 2008) effectively understates the rate of turnover among MPs because of the fraction of MPs who do not seek re-election. Table 1 indicates this averaged 13.2 per cent over the 13 elections and the range was from 3.8 per cent (1980) to 23.2 per cent (1993).
It is useful to examine the percentage of MPs after each election who are “freshmen”. For 13 elections, the average percentage of never-before-elected MPs was 31.7 per cent. The range was from 15 per cent in 2000 to 67.5 per cent in 1993 (Table 1). Thus, on average, almost one-third of MPs coming to Ottawa after each general election have been newcomers (or “freshmen”).
We note that the incumbents’ re-election rates include both one-time and multielection incumbents. For example, Conservative MP Rob Anders (Calgary West, Alta.) has won five consecutive general elections for three different parties.
When he retired just before the 2008 election, NDP MP Bill Blaikie (Elmwood – Transcona, MB) had been in the Commons since 1979, having won nine consecutive general elections. When he retired in 2002, the widely esteemed Liberal MP and former cabinet minister Herb Gray (Windsor West, Ont.) had been in the Commons since1968 1962 having won 13 consecutive elections.
The current dean of the House of Commons, Louis Plamondon (Bas-Richelieu – Nicolet – Bécancour, QC), was first elected as a Conservative in 1984, but now sits as a member of the Bloc Québécois caucus, having won re-election some seven times. Further research is necessary to see if multi-election incumbents have an even greater advantage than one-time ones.
The other side of a high incumbency rate is a low rate of electoral success for challengers. On average, only 7.7 per cent of challengers won a seat in the Commons between 1968 and 2008 (Table 1). Even in 1993, only 10.6 per cent won a seat because there was a huge increase in the total number of candidates (from 1,574 in 1988 to 2,155 in 1993). As we shall see below in Table 2, the election rate for challengers running for a major party is considerably higher.
When challengers were successful, the data in Table 1 indicate that over half the time (54.6 per cent) they beat an incumbent seeking re-election. The range was from 39.3 per cent in 2004 to 81.1 per cent in 1980. The balance of the wins by challengers came from “open seats”—those in which the incumbent was not seeking re-election.
We note that the ratio of total candidates to seats in the Commons increased unevenly from 3.7 in 1968 to 6.0 in 2000 and then declined to about 5.3 in the last three elections. The peak was 7.3 in the unusual election in 1993 in which 14 parties were registered, then a record. (In 2008, 19 parties were registered.)
Re-election rates for incumbents in the United States: a look
Between 1968 and 2006 (19 elections), the re-election rate for incumbents in the House of Representatives was from 85 per cent to 98 per cent. The average was 94.2 per cent, http://www.opensecrets.org).
In 2008, the re-election rate for Democratic incumbents was 97.4 per cent. For Republicans, it was 90.2 per cent. (Both measures take into account incumbents who lost at the primary level.) [source: tabulation from Wikipedia].
Between 1968 and 2006 (19 elections), the average re-election rate for incumbent Senators was lower (81.2 per cent) and it was more varied. The range was from 55 per cent to 96 per cent. The average re-election rate for Senate incumbents between 1968 and 1980 was per cent (Ibid). In 2008, the re-election rate for Democratic Senators was 100 per cent. For Republicans, it was 93.8 per cent (assuming Democrat Al Franken prevails in Minnesota).
Incumbents’ re-election rates by party in Canada, 1988 to 2008
We have calculated the re-election rates of incumbents for the major parties for each of the seven federal general elections between 1988 and 2008. For this purpose, we define a major party as one that elected MPs in that election.
For all elections combined (and note that three parties were in existence for only two or three elections), the party with the highest incumbency rate was the Reform/Canadian Alliance (96.7 per cent) followed very closely by the Conservatives (96.2 per cent), then the BQ (88 per cent), the Liberals (84.4 per cent), and the NDP (67.8 per cent). The Progressive Conservatives trailed at 46.7 per cent, largely due to the 1993 election (Table 2).
Note that for “All Others” (which includes independents and no-affiliation), there were very few incumbents (only 15 over seven elections), and five were re-elected.
There was very little variation across elections in the re-election rates for incumbents for the Conservative Party and the Reform/Canadian Alliance. In contrast, the re-election rate for NDP MPs ranged from 22.2 per cent (1993) to 100 per cent in 2006. The range for the BQ was from 81.4 per cent in 1997 to 100 per cent in 2004 (Table 2).
Table 2 also contains the election rate for challengers by party, by election. Note that this calculation combines two situations. In the first, the party’s incumbent does not run again, but the party retains the seat. In the second, the party’s candidate wins a seat away from another party’s incumbent.
There was a large variation across parties in the challenger election rate when the data are combined for all five elections. Among the major parties, it ranged from 3.7 per cent for the NDP to 45.2 per cent for the BQ. For the Liberals, it was 20.9 per cent, and for the new Conservative Party it was 17.9 per cent. For the “All Others” category (which includes independents and non-affiliated candidates) the election rate for challengers was only 0.08 per cent (three out of 3,720 candidates).
The Conservatives had the least variation across elections: 16.6 per cent to 19.2 per cent for three elections. For the Liberals, the challenger election rate varied from 3.7 per cent (2006) to 46.6 per cent (1993). For the BQ, the range was 15.4 per cent (2000) to 70.6 per cent (1993). More detail can be found in Table 2.
So much for the basic data. We need another longer paper to explain why the incumbency rate for MPs should be so high given the nature of politics and the financing of candidates in Canada.
---------------------------
Table 1: Data on Federal General Elections from 1968 to 2008
Table 2: Re-election Rates for Incumbents and Election Rates for Challengers by Party for Seven Federal General Elections, 1988 to 2008
Incumbency has its benefits, according to elections between 1968 to 2008
A look at the odds of incumbent MPs being re-elected versus the odds of challengers being elected based on the 13 federal general elections between 1968 and 2008.By ALICE FUNKE and W. T. STANBURY
Recent news stories about conflicts within a few Conservative ridings over the party’s new incumbency protection policy, and about new preconditions being imposed on Liberal incumbents to have their nominations protected, once again implicitly raise the question of the electoral benefits of incumbency (see Abbas Rana, The Hill Times, March 16, 2009; Don Martin, National Post, March 27, 2009; Trevor Howell, Calgary Herald, March 29, 2009; and David Akin, Blog, April 6, 2009). The purpose of this piece is to provide data on the odds of incumbent MPs being re-elected versus the odds of challengers being elected, based on the 13 federal general elections between 1968 and 2008. Comparisons are made to the U.S.
Political party policies for incumbent MPs and nomination contests
In The Hill Times of March 16, 2009, Abbas Rana provided a convenient summary of the current policies of three major parties concerning nomination contests for incumbent MPs
New Democratic Party: “has a policy of not protecting their incumbent MPs from the nomination process and requires MPs to win open nomination contests prior to every federal election.”
Conservatives: “Riding association members can ask us for a race and if twothirds or more of the membership in any riding asks for a nomination race then there will be a contest.” (Riding members receive a ballot in the mail to decide whether to hold a nomination contest.)
Liberals: Incumbents must show to the central party headquarters by June [2009] that they have 400 members in their riding associations and at least 40 “Victory Fund” donors who donate $10 or more on a monthly basis, $5 to the party headquarters, and the other $5 to the riding association.
Bloc Québécois: According to La Presse, the Bloc started its own process of 75 nomination meetings last week, and expects to have 40/75 candidates nominated by September (Joël-Denis Bellavance, April 20, 2009)
Party policies have varied over time. Andrew Steele (blogpost in The Globe and Mail, April 7, 2009) pointed out that “In 2006, the [Conservative Party] did not allow any contest against sitting MPs. In 2008, the window for contesting nominations was peculiarly short.”
Further, “the Liberals automatically renominated all incumbents in the 2000 election, citing electoral urgency because the Prime Minister wanted to call an election.”
John Ivison (National Post, March 10, 2009) notes that “at the Liberal National Convention… in 1993, Jean Chrétien rammed through a resolution that deprived grass-roots Grits of the only power they had—the ability to choose their candidate during a nomination process…. Mr. Chrétien said he would protect the nomination of MPs indefinitely, provided they did not speak out against him.”
Re-election rates for incumbent MPs between 1968 to 2008
To examine incumbency rates (the percentage of incumbent MPs seeking reelection who were re-elected) we gathered the data on 13 general elections between 1968 and 2008 (Table 1). The 13 elections resulted in five different types of government in Ottawa: six Liberal majorities; two Liberal minorities; two Progressive Conservative majorities (under Brian Mulroney); one PC minority and two Conservative Party minorities (under Stephen Harper). Thus five of 13 elections resulted in a minority government over the 40-year period with three since 2004.
The overall rate of re-election for incumbent MPs was 78.1 per cent. In only two of 13 elections (1984 = 60.2 per cent; 1993 = 41.1 per cent) did the incumbency rate fall below 70 per cent. For seven elections, the rate was over 80 per cent, and in 2000 it was just over 90 per cent (Table 1).
The low incumbency rate in 1993 (41.4 per cent) reflected a most unusual election. This first of three majority Liberal governments under Jean Chrétien saw the Liberals gain 177 seats from 83 in 1988. More importantly, two new parties had a big breakthrough. The recently-created Bloc Québécois, running a full slate of 75 candidates in Quebec won 54 seats. Thus a separatist party became the official opposition.
The Reform Party, created in 1987, went from zero seats in 1988 to 52 in 1993. The NDP collapsed from a record 43 seats in 1988 to only nine in 1993. However, the Progressive Conservatives under Kim Campbell (who had replaced Brian Mulroney) were almost wiped out, falling from 169 seats in 1988 to two in 1993. Thus five parties obtained seats in the Commons in 1993 (Table 1). That election also saw the lowest percent of incumbents seeking reelection in the period 1988 to 2008 (220 of 295 = 74.6 per cent), as compared with the high watermark of 2006 (278 of 308 = 90.3 per cent).
Evidence that 1993 was an anomaly can be found in the next election (1997) when the Liberals won only 155 seats (a drop of 22) in an enlarged House of 301 seats, but 82.6 per cent of Liberal incumbents were re-elected.
The other year in which the incumbency rate was low (60.2 per cent in 1984) also reflected highly unusual conditions. Brian Mulroney won the greatest sweep in Canada’s political history. The PCs won 211 of 282 seats while the Liberals won only 40. In the earlier 1980 election, the PCs had won only 103 seats compared to the Liberals 147. This number of first-time MPs in 1984 (133 of 282 = 47.2 per cent) was the second highest in the 13 elections reported in Table 1. The highest was in 1993: 199 of 295 or 67.5 per cent.
The high incumbency rate (an average of 78.1 per cent between 1968 and 2008) effectively understates the rate of turnover among MPs because of the fraction of MPs who do not seek re-election. Table 1 indicates this averaged 13.2 per cent over the 13 elections and the range was from 3.8 per cent (1980) to 23.2 per cent (1993).
It is useful to examine the percentage of MPs after each election who are “freshmen”. For 13 elections, the average percentage of never-before-elected MPs was 31.7 per cent. The range was from 15 per cent in 2000 to 67.5 per cent in 1993 (Table 1). Thus, on average, almost one-third of MPs coming to Ottawa after each general election have been newcomers (or “freshmen”).
We note that the incumbents’ re-election rates include both one-time and multielection incumbents. For example, Conservative MP Rob Anders (Calgary West, Alta.) has won five consecutive general elections for three different parties.
When he retired just before the 2008 election, NDP MP Bill Blaikie (Elmwood – Transcona, MB) had been in the Commons since 1979, having won nine consecutive general elections. When he retired in 2002, the widely esteemed Liberal MP and former cabinet minister Herb Gray (Windsor West, Ont.) had been in the Commons since
The current dean of the House of Commons, Louis Plamondon (Bas-Richelieu – Nicolet – Bécancour, QC), was first elected as a Conservative in 1984, but now sits as a member of the Bloc Québécois caucus, having won re-election some seven times. Further research is necessary to see if multi-election incumbents have an even greater advantage than one-time ones.
The other side of a high incumbency rate is a low rate of electoral success for challengers. On average, only 7.7 per cent of challengers won a seat in the Commons between 1968 and 2008 (Table 1). Even in 1993, only 10.6 per cent won a seat because there was a huge increase in the total number of candidates (from 1,574 in 1988 to 2,155 in 1993). As we shall see below in Table 2, the election rate for challengers running for a major party is considerably higher.
When challengers were successful, the data in Table 1 indicate that over half the time (54.6 per cent) they beat an incumbent seeking re-election. The range was from 39.3 per cent in 2004 to 81.1 per cent in 1980. The balance of the wins by challengers came from “open seats”—those in which the incumbent was not seeking re-election.
We note that the ratio of total candidates to seats in the Commons increased unevenly from 3.7 in 1968 to 6.0 in 2000 and then declined to about 5.3 in the last three elections. The peak was 7.3 in the unusual election in 1993 in which 14 parties were registered, then a record. (In 2008, 19 parties were registered.)
Re-election rates for incumbents in the United States: a look
Between 1968 and 2006 (19 elections), the re-election rate for incumbents in the House of Representatives was from 85 per cent to 98 per cent. The average was 94.2 per cent, http://www.opensecrets.org).
In 2008, the re-election rate for Democratic incumbents was 97.4 per cent. For Republicans, it was 90.2 per cent. (Both measures take into account incumbents who lost at the primary level.) [source: tabulation from Wikipedia].
Between 1968 and 2006 (19 elections), the average re-election rate for incumbent Senators was lower (81.2 per cent) and it was more varied. The range was from 55 per cent to 96 per cent. The average re-election rate for Senate incumbents between 1968 and 1980 was per cent (Ibid). In 2008, the re-election rate for Democratic Senators was 100 per cent. For Republicans, it was 93.8 per cent (assuming Democrat Al Franken prevails in Minnesota).
Incumbents’ re-election rates by party in Canada, 1988 to 2008
We have calculated the re-election rates of incumbents for the major parties for each of the seven federal general elections between 1988 and 2008. For this purpose, we define a major party as one that elected MPs in that election.
For all elections combined (and note that three parties were in existence for only two or three elections), the party with the highest incumbency rate was the Reform/Canadian Alliance (96.7 per cent) followed very closely by the Conservatives (96.2 per cent), then the BQ (88 per cent), the Liberals (84.4 per cent), and the NDP (67.8 per cent). The Progressive Conservatives trailed at 46.7 per cent, largely due to the 1993 election (Table 2).
Note that for “All Others” (which includes independents and no-affiliation), there were very few incumbents (only 15 over seven elections), and five were re-elected.
There was very little variation across elections in the re-election rates for incumbents for the Conservative Party and the Reform/Canadian Alliance. In contrast, the re-election rate for NDP MPs ranged from 22.2 per cent (1993) to 100 per cent in 2006. The range for the BQ was from 81.4 per cent in 1997 to 100 per cent in 2004 (Table 2).
Table 2 also contains the election rate for challengers by party, by election. Note that this calculation combines two situations. In the first, the party’s incumbent does not run again, but the party retains the seat. In the second, the party’s candidate wins a seat away from another party’s incumbent.
There was a large variation across parties in the challenger election rate when the data are combined for all five elections. Among the major parties, it ranged from 3.7 per cent for the NDP to 45.2 per cent for the BQ. For the Liberals, it was 20.9 per cent, and for the new Conservative Party it was 17.9 per cent. For the “All Others” category (which includes independents and non-affiliated candidates) the election rate for challengers was only 0.08 per cent (three out of 3,720 candidates).
The Conservatives had the least variation across elections: 16.6 per cent to 19.2 per cent for three elections. For the Liberals, the challenger election rate varied from 3.7 per cent (2006) to 46.6 per cent (1993). For the BQ, the range was 15.4 per cent (2000) to 70.6 per cent (1993). More detail can be found in Table 2.
So much for the basic data. We need another longer paper to explain why the incumbency rate for MPs should be so high given the nature of politics and the financing of candidates in Canada.
---------------------------
Table 1: Data on Federal General Elections from 1968 to 2008
Table 2: Re-election Rates for Incumbents and Election Rates for Challengers by Party for Seven Federal General Elections, 1988 to 2008
Labels: Incumbency
posted by The Pundits' Guide at
9:28 AM
2 Comments:
Good article. But I have to wonder whether looking at re-election rates alone would result in a lack of distinction between party loyalty and individual incumbency (particularly in the wake of an election where the only three party-jumpers who ran again lost their seats) - is there enough data to control for that difference?
I'll take a look, and see if I can get back to you with the answer. We had originally broken down the challengers between those who won seats previously held by the party, from challengers who gained seats from other parties, but the article was starting to get a bit too long.
I'll post the data in the next day or so.
Post a Comment
Links to this post:
Create a Link
<< Home