In Beyond Presidentialism and Parliamentarism (Oxford University Press, 2021), Steffen Ganghof attempts to overcome the alternative between presidentialism and parliamentarism by arguing for ‘semi-parliamentarism.’ According to Ganghof, it is true that presidentialism, contrary to (pure) parliamentarism, separates the executive and the legislative power. However, the executive power is concentrated in one person and this ‘executive personalism’ (p. 1, p. 15) tends to strengthen presidential power and weaken the presidential separation of powers, as demonstrated by Donald Trump [and Vladimir Putin at this moment]. To avoid this, Ganghof defends the idea of ‘semi-parliamentarism’: while semi-presidentialism divides executive power into two, semi-parliamentarism divides legislative power into two parts, thus institutionalizing a ‘separation of powers between the executive and one part of a directly elected assembly’ (p. 108. Regarding the definition of semi-parliamentarism, see p. 50).
In the version of a ‘single-vote, mixed-member’ system, the first part (confidence seats) of the assembly has the power of (no-)confidence. The assembly as a whole (confidence seats plus legislative top-up seats) has the power of ‘legislation, deliberation, and control’ (p. 142). The members of the assembly are directly elected by the run-off or modified AV system with a single at-large district. After all seats are distributed to parties in proportion to the first (preference) votes each party polls, the confidence seats are distributed to the top two parties in proportion to the final votes these parties poll after the run-off election or the transfer of votes. Thus, semi-parliamentarism divides the assembly into two parts to institutionalize the separation of powers without executive personalism.
There is no doubt that this book is highly stimulating for normative political theorists. While it contains many stimulating points for discussion, such as the replacement of the framework of majoritarian vs. consensus democracy with the framework of simple vs. majoritarianism (Chapter 5), I will concentrate on examining his institutional design of semi-parliamentarism presented in Chapter 8 (Note that Ganghof states that the ‘discussion was necessarily explorative, selective, and preliminary.’ p. 147).
I believe that the most serious problem is that the institutional design presented in Chapter 8 is not suitable for the conception of semi-parliamentarism. If the two-round or modified AV system is adopted for electing the confidence seats, parties will seek a pre-electoral coalition, whether weaker (Table 8.1) or stronger (Table 8.2) to win the confidence seats election. If two pre-electoral coalitions are formed, voters can identify not only the prime minister but also the governing parties and their policies. However, a separation of power must be undermined by the fusion of executive and legislative powers because the parliament is under the control of the governing coalition.
To avoid this unexpected fusion of powers, a non-electoral design of semi-parliamentarism is needed. What should be noted is that Ganghof refers to Abizadeh’s conception of semi-parliamentarism, which uses sortition as well as election (p. 47, p. 58). However, he does not pursue the use of sortition to institutionalize semi-parliamentarism. However, there must be a sorted House of Citizens as well as an elected House of Representatives to realize a semi-parliamentary separation of powers between the executive power and one of the legislative powers.
Suppose that the parliament consists of two houses: the sorted House of Citizens and the elected House of Representatives. Since members of the former are selected by sortition, the House of Citizens is independent from the government as well as the House of Representatives. In addition, the independence will be strengthened if some strong powers are given to the House of Citizens: the veto power to reject the decisions of the House of Representatives and the recommendation power to urge the House of Representatives to make a law (岡﨑晴輝「選挙制と抽選制」、『憲法研究』第5号、2019年11月、87-96頁。http://hdl.handle.net/2324/2557147). I would like to add to this review the third power: no-confidence power to dismiss the prime minister and the cabinet (since this power is a deviation of Ganghof’s definition of semi-parliamentarism on p. 50, it is necessary to revise the definition of semi-parliamentarism). These powers will guarantee a substantial separation of powers between the executive power and the House of Citizens.
Since the existing elected House of Councillors is replaced by the sorted House of Citizens, the House of Representatives must undertake a difficult task to achieve the ideal of identifiability and accountability and the ideal of proportionality at once. It is true that the modified AV system that Ganghof proposed is an option. However, the system is weak in reconciling the ideal of identifiability and accountability and the ideal of proportionality (cf. OKAZAKI Seiki, “The Second Defense of Proportional Representation with a Limited Majority Bonus,” Seiji Kenkyu, No. 68, March 2021, pp. F1(130)-F19(112). http://hdl.handle.net/2324/4377858).
Instead, I argue that the members of the House of Representatives are elected by the proportional representation system with a limited majority bonus (PR-LMB). It is a PR system that gives 55 percent of the seats to the winning coalition or party and 45 percent to the other parties. Under the PR-LMB, voters can choose parties in the parliament because the PR element distributes seats proportionally except for a limited majority bonus. Voters can also choose the government because the LMB element encourages the formation of a bi-coalitional competition. Thus, PR-LMB makes the ideal of identifiability and accountability and the ideal of proportionality compatible in the House of Representatives (OKAZAKI Seiki, “In Defense of Proportional Representation with a Limited Majority Bonus,” Hosei Kenkyu, Vol. 85, No. 3-4, March 2019, pp. F1(1516)-F19(1498). http://hdl.handle.net/2324/2231036).
The conclusion of this review is very simple: (1) the sorted House of Citizens is indispensable for institutionalizing semi-parliamentarism, and (2) the members of the House of Representatives should be elected by PR-LMB in order to achieve the ideal of identifiability and accountability and that of proportionality at once. I believe that this mixed bicameralism is more suitable for the conception of semi-parliamentarism than Ganghof’s design. I hope that Professor Ganghof will respond to these arguments in his next article.
OKAZAKI, Seiki (Kyushu University, Japan)
(Political Theory Forum, 26th February 2022)
