The last decade has seen multi-party competition for elected legislatures initiated or expanded in Tunisia, Algeria, Morocco, Egypt, Jordan, Yemen, Kuwait, Lebanon and the Palestinian Authority. Executive authority in most cases remains an uncontested, if not completely unelected, post. Nevertheless, incumbent rulers invariably tout these legislative elections as evidence of domestic legitimacy, often anointing their countries as “on the road to democracy” in their wake.
The evidence to the contrary is well known: The powers of the judicial and legislative branches are sharply constrained, opposition parties have only limited access to media and are usually restricted in their campaigning activities, new political parties must receive government permission to become legal, votes in support of the ruling party are often coerced, and various means of falsifying the actual vote count (tazwir) obtain. None of these elections have changed governments, or even altered unpopular policies. These are elections that do not disturb their authoritarian milieu.
Nonetheless, these political openings constitute a departure from past practices and contrast with some neighboring states, where legislatures are either non-existent or chosen without contestation. Holding contested elections concedes that citizens have a right to self-selected political representation. Even carefully controlled multi-party elections can provide a limited forum for public debate about society’s future and limited opportunities for mobilization and freedom of expression, and an opposition presence in parliament provides a platform for regime critics to promote their views.
This position contrasts with those that deny the agency of secular, and non-violent Islamic, parties and groups,  as well as those that assume political openings will inevitably spawn democracy.  The future is uncertain, and contingent on politics.
Here I focus on just one aspect of these elections: the politics surrounding electoral systems. Electoral design involves the drawing of district lines, the number of representatives elected per district and the rules which determine the winner. Different systems can produce vastly different legislatures, with implications for policy outcomes. They also affect the number of viable political parties in a country and the means by which they campaign. Manipulating electoral systems thus provides authoritarian elites with a subtle mechanism for controlling immediate electoral outcomes and partisan politics more broadly.
Winner Takes All
Most contested Arab elections use winner-takes-all systems, among whose consequences is the distortion of the actual vote given to large, well-known parties. Egypt’s 1976 parliamentary elections were held under a majority runoff rule. Of 342 available seats, the self-proclaimed centrist “platform,” representing Sadat supporters, won 280, or more than 80 percent, but with only about 60 percent of the popular vote.  The same effect was evident in Palestine’s 1996 legislative council elections, which used an at-large system designed by an Arafat-appointed commission. Candidates running on a Fatah slate received only 30 percent of the votes, but wound up with 58 percent of the 88 seats. Conversely, independents (many of whom were Fatah activists who had not been selected for the Fatah slate in their districts) polled about 60 percent of the total vote, but received only about 40 percent of the seats. 
In the Palestine example, Fatah’s position was also enhanced by the fact that rival factions within the PLO, as well as many Islamists affiliated with Hamas, refused to participate in the elections, charging that this would imply acceptance of the Oslo accords. But as Khalil Shikaki noted, the smaller parties’ decision to boycott was encouraged by the choice of an electoral system that gave them little opportunity to capture seats. Those opposition parties which did participate received 10 percent of the total votes, but only three percent of the seats. Before the draft election law was approved, several opposition parties as well as the Geneva-based International Commission of Jurists had urged the adoption of a proportional system in order to give greater voice to opposition parties. 
Ironically, the same at-large system that privileged Fatah worked to King Hussein’s disadvantage when used in Jordan’s 1989 elections. Because a 32-year ban on political parties remained in effect, the “national recognition” advantage went to the Muslim Brotherhood, who were able to sponsor candidates as a charitable organization. Despite gerrymandering that gave more representation per person to East Bank districts presumed to be more pro-regime, Brotherhood and Islamist-identified candidates won about 42 percent of the 80 contested seats, a considerably greater proportion of seats than the percentage of votes they polled. 
The effects of electoral designs that favor incumbents are predictably overlooked in Western coverage of elections. A good example was Jordan’s 1993 elections. The press rushed to hail the results after King Hussein retooled the election system to “correct the problem” of the 1989 poll. A year earlier, the monarchy had permitted the formation of political parties, and some 24 of them, many reflecting pan-Arab, pan-Islamic or Palestinian nationalist views, had been legalized before the poll. But the king also decreed a switch to a “limited vote” system. As the monarchy anticipated, many citizens awarded their single vote to a tribal favorite rather than their ideological preference. The resulting parliament had a sharply diminished opposition presence. The Islamic Action Front, a coalition dominated by the Brotherhood, took 16 seats, and independent Islamists another four, so Islamist forces now held only 25 percent of the seats. Another eight seats were captured by various leftists, leaving roughly a 65 percent majority to generally pro-regime figures. Thus the king got the compliant legislature he wanted while appearing to promote democratization, and both the regime and its Western backers trumpeted the results as proof that Jordanians backed Hussein’s foreign policies: “It would appear His Majesty’s call for moderation has been answered,” the state-controlled Radio Jordan declared,  while the New York Times proclaimed that “Jordanians have given a strong endorsement to the Middle East peace effort by spurning Islamic militants whose principal platform was opposition to any peace talks with Israel.” 
Egypt’s Changing Election Laws
Egyptian President Husni Mubarak resorted to election re-engineering at the same time he increased freedoms for opposition parties. Mubarak was apparently unsatisfied with the healthy majority that the previous election law had delivered to the ruling party. In the 1976 poll, most of the seats that government supporters did not capture went to independents. Mubarak’s 1984 election law closed the door on independent candidacies, while limiting the chances for opposition parties. This law was eventually overturned by legal challenges. The examples above illustrate the importance to opposition groups of struggling over election laws. The Egyptian case also highlights the need for careful consideration of an alternative. Egypt’s new election law seems more disadvantageous to opposition parties than the one it replaced.
Mubarak initially chose a party-list proportional representation system, but with such a high threshold — 8 percent — that it negated the system’s advantages for most opposition parties. The law was particularly onerous because it required that a party poll 8 percent of the vote nationally in order to win seats in any particular constituency. Moreover, small parties’ votes that could not be used to obtain seats would accrue to the largest party, i.e., the National Democratic Party (NDP), instead. 
Despite all these advantages, the NDP still resorted to considerable interference and fraud, and Mubarak refused to subject elections to supervision by a judicial body. Final results showed the NDP winning 87 percent of the contested seats, but only 73 percent of the reported popular vote. The gap between the ruling party’s reported popular vote and its more substantial majority in the legislature reflects the NDP’s acquisition of votes cast for the leftist Tagammu‘ and an Islamist-nationalist rival, the Socialist Labor Party, both of which failed to reach the threshold. An alliance of the New Wafd Party and the Muslim Brotherhood was the only opposition grouping able to surmount the 8 percent barrier, obtaining the remaining 13 percent of the seats.
The 1984 election law was changed after a court challenge by opposition lawyers. Three years later, as it became apparent that the Supreme Court was about to rule in their favor, the regime preempted the court by disbanding the parliament and issuing a new electoral law. It set aside one seat in each constituency for independent candidates, although parties were also permitted to nominate individuals for these seats. They were contested on a plurality basis, with a required runoff if the top candidate did not poll more than 20 percent of the vote. For the rest of the seats, the 8 percent threshold proportional representation system remained in effect. The creation of individually contested seats provided an opportunity for the smaller opposition parties to obtain seats by running locally popular candidates as independents. It was also an opening for parties denied legal recognition by the government and activists disillusioned with both the legal and clandestine parties available. Yet, seats were also sought by NDP loyalists denied positions on the district party list. Thus, most districts showed 40-50 names on the ballot. 
The 1987 election did produce a more diversified legislature, and while the NDP still enjoyed a comfortable margin above the two-thirds majority needed to pass legislation and ensure Mubarak’s renomination, its hold had fallen to 79 percent. Ironically, though, the party fared somewhat better in the competition for individual seats than in the proportional poll. With near 70 percent of the party-list vote, it captured 309 out of 400, or 77 percent, of those seats. Yet NDP-sponsored independents won 39 out of 48, or 81 percent, of the individual seats, with another five won by NDP members who ran without their party’s backing.
In May 1990, Egypt’s Supreme Constitutional Court ruled that the 1986 election law unfairly discriminated against independents, and declared the parliament elected under the law to be “null and void.” Mubarak suspended the parliament — two years early — and appointed a committee of legal experts to draft a new elections law.
Although the new law restored the majority runoff system, it failed to guarantee independent supervision of the elections. For this reason, all of the established legal opposition parties except Tagammu‘ boycotted the elections in December 1990. Among the unofficial parties, the Green, Nasserist and Egyptian Communist parties participated, while the Muslim Brotherhood and some smaller Communist groups joined the boycott. All told, about 3,000 candidates competed, roughly seven for each seat, including 789 who were affiliated with the NDP but ran without their party’s backing. 
Weakening Political Parties
The election results were telling: NDP-backed candidates won only 253 seats, with some significant defeats for party secretaries in six governorates. However, 95 of the party affiliates who ran as independents won, and most of these were quickly coopted into the party’s parliamentary committee. Thus, the NDP could again count on more than 75 percent of the legislature’s votes. As far as parliamentary control is concerned, a winner-take-all system worked as well for the ruling party as had the previous system. The individual candidate process did, however, prevent party leaders from managing which NDP members would sit in the legislature.
The new system promoted dissension within some opposition parties as well. The boycott decisions were controversial, and numerous opposition figures defied their parties and ran as independents. Thus, the new system provided campaign opportunities for true independents and members of illegal parties at the expense of the discipline and cohesion of the legal opposition. This phenomenon may have contributed to the violence that marred Egypt’s 1995 parliamentary elections.  Several new parties had been legalized, and the government agreed to double the limited amount of airtime available to the opposition. That, and promises by Mubarak and his interior minister that the voting would be free and fair, led all opposition groups to participate. The number of candidates approached 4,000, almost ten per seat. But opposition leaders and human rights activists reported an unprecedented level of repression and interference. The NDP substantially increased its majority in spite of enhanced contestation. Some charge that this was because party leaders questioned the loyalty of members who ran as independents when denied the party’s backing. 
At the outset, the government launched a crackdown on the Muslim Brotherhood, arresting over 700 and prompting some candidates associated with the organization to withdraw from the campaign. Supporters of those who remained in the race and other opposition candidates were subject to intimidation and violence at the hands of the security forces on polling day, while police assisted government agents in stuffing ballot boxes. In addition, armed conflicts erupted between supporters of rival candidates in numerous areas, attributed by some observers to longstanding rivalries among elites. All told, about 50 people were killed and almost 900 wounded.
In the election’s first round, there were 138 winners, of whom 124 were NDP-backed; 11 of the 14 victorious independents were NDP affiliates. No opposition party candidates captured seats. NDP-selected candidates then claimed 183 of the 306 seats which went to runoff, while four opposition parties split 13. Most of the winning independents were again NDP loyalists, with half declaring sympathy for the Islamists. Thus, the ruling party emerged from the poll with 94 percent of the contested seats.
The shift away from proportional representation in Egypt thus appears to have weakened opposition parties, or at best failed to help them. Even without the heightened violence and repression, the winner-take-all system would have enhanced advantages enjoyed by the ruling party. A proportional representation system with a more reasonable threshold — in the range of 4 percent — combined with more democratic selection processes within opposition parties, would serve much better to weaken the NDP’s hold on the legislature.
Rocking the Boat
The mechanism of electoral manipulation described here has been largely ignored by Western scholarship and media. Electoral system engineering seems likely to increase in direct proportion to the success of domestic and/or foreign election monitors’ ability to prevent the time-honored practices of interference and fraud. Government challengers can make gains by engaging authoritarian regimes in debate and struggle over these “hidden” institutional arrangements.
Which electoral system is best for a given democracy is the subject of legitimate debate, although winner-take-all clearly operates against minority representation and favors established national parties. The calculus must be different in any case for nondemocratic countries. Proportional representation systems, which increase the representation of opposition forces, would better serve to undermine incumbent authoritarian regimes. Fighting for proportional representation is also an arena in which opposition parties can cooperate in spite of ideological differences, because all stand to gain seats if the demand is won. Of course, ruling incumbents would strongly resist any fairly structured proportional system, but exposing their recalcitrance would further erode the legitimacy of regimes’ purported commitment to genuine political representation.
Author’s Note: The author thanks Eva Bellin, Steve Heydemann, Ann Lesch, Roger Owen, Rob Richie and several Middle East Report editors for helpful comments on earlier drafts.
 Ghassan-Salame, “Introduction: Where are the Democrats?” and John Waterbury, “Democracy Without Democrats? The Potential for Political Liberalization in the Middle East,” in Salame, ed., Democracy Without Democrats: The Renewal of Politics in the Muslim World (New York: I. B. Tauris, 1994).
 Augustus Richard Norton, ed., Civil Society in the Middle East (Leiden: E. J. Brill, 1995) and Saad Eddin Ibrahim, “Overview,” in Rex Brynen, Bahgat Korany and Paul Noble, eds., Political Liberalization and Democratization in the Arab World (Boulder, CO: Lynne Rienner, 1995).
 Mark Cooper, The Transformation of Egypt (Baltimore: Johns Hopkins University Press, 1982), pp. 204-234.
 Many independents were Fatah activists who had not been selected for the Fatah slate in their district. Jerusalem Post, January 23, 1996; Deutsche Press Agentur, January 22, 1996; As‘ad Ghanem, “Founding Elections in a Transitional Period: The First Palestinian General Elections,” Middle East Journal 50/4 (Fall 1996).
 Ghanem, op cit; Jerusalem Post, June 16, 1995, December 7, 1995 and January 19, 1996.
 On Jordan’s 1989 and 1993 elections, see Middle East Magazine, October 10, 1989; Kamel S. Abu Jaber and Schirin H. Fathi, “The 1989 Jordanian Parliamentary Elections,” Orient 31 (1990); Andrew Reynolds and Jorgen Elkit, “Jordan: Electoral System Design in the Arab World,” International IDEA Handbook of Electoral System Design (Stockholm: International Institute for Democracy and Electoral Assistance, 1997); Middle East Economic Digest, November 24, 1989; Timothy Piro, “Parliament, Politics and Pluralism in Jordan,” Middle East Insight (July-October 1992); Tim H. Reidel, “The 1993 Parliamentary Elections in Jordan,” Orient 35 (1994), p. 53; Ben Wederman, “Democracy in Jordan,” Middle East Insight (November-December 1993); and MEED, November 19, 1993.
 United Press International, November 9, 1993.
 New York Times, November 10, 1993.
 Another aspect of the law designed to disadvantage small parties required them to nominate enough candidates to fill every available seat, as well as designating alternates who could replace an elected official in the case of death or resignation; the total number of candidates needed was 690. See Fauzi Najjar, “Elections and Democracy in Egypt,” American-Arab Affairs (Summer 1989), and Bertus Hendriks, “Egypt’s Elections, Mubarak’s Bind,“ MERIP Reports 129 (January 1985).
 1987 election information from Erika Post, “Egypt’s Elections,” Middle East Report 147 (July-August 1987), and personal observation.
 1990 election information from Gehad Auda, “Egypt’s Uneasy Party Politics,” Journal of Democracy 2/2 (Spring 1991); Africa Economic Digest, May 28, 1990; Middle East Economic Digest, September 28 and October 12, 1990; Guardian, November 29, 1990; and an interview with Ahmad Abdalla, April 1993.
 On the 1995 elections, see Mona Makram-Ebeid, “Egypt’s 1995 Elections: One Step Forward, Two Steps Back?” Middle East Policy 4/3 (March 1996); al-Ahram Weekly, December 7-13 and December 14-20, 1995; Reuters, UPI and Deutsche Press Agentur, November-December 1995; and Reuters, October 4, 1996.
 Eberhard Kienle, “More than a Response to Islamism: The Political Deliberalization of Egypt in the 1990s,” Middle East Journal 52/2 (Spring 1998).