Canada

encing virtually no retrenchment since 1983, except for an increase in taxation of Social Security benefits for upper-income recipients in 1993.

A very broad range of restructuring options for Social Security has been discussed over the past decade. The Clinton administration proposed allowing investment of Social Security surpluses in equities markets to increase returns, in addition to a system of supplemental individual retirement sayings accounts. At the beginning of his second term, George W. Bush proposed a major initiative to allow individuals to partially opt out of Social Security into a system of funded individual accounts. Even though Bush has significant majorities both chambers of Congress, the political sensitivity of Social Security and the absence of an immediate funding crisis make passage of a major reform package very much an uphill battle.

Canada

Like the United States, Canada confronts a demographic challenge that is more modest than that in most other industrialized countries. However, Canada also confronted severe fiscal pressures in the 1980s through the mid-1990s that made social spending politically vulnerable. Canada's public pension system is multitiered, with a quasi-universal (phased out at high incomes) flat-rate Old Age Security program and an income-tested Guaranteed Income Supplement, both financed by general revenues. In addition, there is a contributory, earnings-related Canada Pension Plan (Quebec operates a parallel and integrated Quebec Pension Plan) on top of OAS that is financed entirely by payroll taxes.

In general, Canada's Westminster-style parliamentary system facilitates making politically contentious decisions because there is no separation of executive and legislative power and the second legislative chamber (the appointive Senate) is very weak. Moreover, Canada had uninterrupted sin- gle-part majority government from 1980 until 2004 and has had weak and divided opposition parties since 1993. However, the structure of the CPP/QPP inhibits policy change because revisions must be approved by the federal government and a supermajority of provinces.

Pension-policy change in Canada, as in the United States, has been limited. A benefit "clawback" for upper-income recipients of Old Age Security was introduced by a Conservative government in 1988, using procedurally privileged budget legislation that is almost impossible for opposition parties to block or amend so long as the government has a majority, but only after the same government backed down from a 1985 initiative to partially de-index OAS benefits. And Canada has experienced only very modest restructuring of its public pensions in recent years. In the mid-1990s, the Chretien government announced plans to merge the quasi-universal OAS program and the income-tested GIS in a way that would further reduce benefits for upper-income recipients. But the plan was repeatedly delayed because of its political sensitivity, and it was eventually shelved when emerging federal budget surpluses made cutting public pensions seem like a political nonstarter.

For the payroll-tax-financed Canada Pension Plan, efforts to address growing deficits in the program were initially delayed by the difficulty of securing the broad federal-provincial agreement needed to revise the program. But, once Ottawa and most provincial governments reached an agreement in 1997, there was little that opponents could do to block or revise it. The outcome was a major contribution-rate increase enacted in 1997, but only modest (and hard-to-see because of their technical nature) cutbacks in benefits.²² But a broader restructuring toward individual accounts as a mandatory component of or opt-out from the CPP made even less headway than in the United States; the major opposition party favored such a shift but was powerless to get it on the agenda in a political system where the governing party has a virtual monopoly on the policy agenda.

United Kingdom

~A quasi-universal flat-rate basic pension has long been the anchor of the United Kingdom's public pension system. A prolonged stalemate between Labour and the Conservatives delayed adoption of a State Earnings Related Pension Scheme (SERPS) until 1975. Because many workers were already covered by occupational pension schemes by this point, SERPS included an opt-out for approved occupational schemes rather than keeping them as an add-on to the state scheme. The United Kingdom also relies heavily on income-tested benefit programs for the elderly.

Several aspects of U.K. political institutions facilitate both short-term loss imposition and policy regime change: the lack of separation of executive and legislative power, single-party majority government since the 1979 election (frequently-with very large majorities in the House of Commons), and extremely weak bicameralism.

• ^Policy outcomes in the U.K retirement-income system reflect this concentration of power; there has been major restructuring as well as frequent

 

tinkering. Under Margaret Thatcher, indexing of the Basic State Pension shifted to prices from the higher of wages and prices, resulting in very large long-term cuts in benefits. The Thatcher government also marginalized SERPS by lowering benefits and creating additional incentives to opt-out into portable "personal pensions." Not surprisingly, this restructuring was enacted by a government that enjoyed strong formal powers, a clear majority in the House of Commons, and an exceptionally weak and divided Labour opposition. Even the Thatcher government was not omnipotent, however; it had to back away from its initial 1985 proposal to abolish SERPS entirely in the face of Widespread criticism from its usual allies (employers and pension providers) and a pledge from Labour Party leader Neil Kin- nock to reverse the change when Labour came back into power.²³

Additional, generally incremental changes have been made under Tony Blair's New Labour. SERPS has been converted to the State Second Pension (S2P), which will increase benefits for low lifetime earners but is also intended to lower replacement rates for high earners in the long run. Other changes under Blair have generally been modestly expansionary, while addressing problems in Britain's unusual public-private pension mix. These include a series of ad hoc increases in the Basic Pension, the introduction of "stakeholder" pensions to provide an alternative to high and opaque personal pension fees, a new Pension Credit to give low earners greater incentives to save for retirement, and recent proposals for pension insurance for private pensions.

New Zealand

Like the United States and Canada, New Zealand faces a less severe aging crisis than most European countries in the short and medium terms. But New Zealand has also encountered relative economic decline and repeated budgetary and economic crises over the past thirty years.

Policy feedbacks have also heavily influenced policy choices. At the beginning of the 1970s, New Zealand relied on a combination of a universal flat-rate pension payable at age 65 and a means-tested pension, payable at age 60, both paid from general revenues. The 1972-1975 Labour government sought to add a payroll-tax-financed earnings-related pension tier, but the National government elected in 1975 quickly abolished it and substituted a more generous universal pension financed by general revenues. The absence of a trust-fund device has meant that there is no action-forcing and legitimizing mechanism for pension retrenchment, while flat-rate

 

 

pensions paid to all have made it more difficult to impose retrenchment gradually by "grandfathering" current recipients and cutting benefits for later ones.

New Zealand's political institutions have had complex and ambiguous impacts. Until 1996, New Zealand's unicameral legislature was elected in single-member constituencies, leading to consistent single-party majority governments—although not always the party that won a plurality of the popular vote. Since the adoption of Mixed-Member Proportional (MMP) electoral rules in 1996, New Zealand has consistently had coalition governments. But continuation of some Westminster traditions (including quick passage of legislation through use of "urgency" procedures) even after onset of MMP in 1996 continue to concentrate both governmental power and governmental accountability. Moreover, the combination of short electoral cycles (generally three years) and close electoral competition ^ gives politicians strong incentives to act quickly and with minimal consultation to try to embed policy and allow long-term gain to emerge from short-term pain before next election. But short electoral cycles also make 1: it likely neither that policies will be embedded nor losses forgiven by next Infection.

l|g|^^;Changes in public pension policy in New Zealand over the past two x decades reflect these patterns. A National Party government was able to / enact and implement a rapid increase in the age of eligibility for New i^S&aland Superannuation in the early 1990s, but efforts to make substantial r short-term cuts in Superannuation benefits relative to the average wage fg§|pn&~to income-test benefits for upper-income recipients proved more ^■■(difficult to sustain. Efforts to limit politicking over pensions through a ipiiffltiparty accord in the early 1990s foiled to survive the move to MMP in ЯШШ ^w^^{en a new} "swing" party, New Zealand First, condemned the IpSccord and called for a restoration of many of the cuts made in the first '• part of the decade. As a junior coalition partner in the 1996-99 govern- ptfftent, NZ First also insisted on a referendum on a complex new individ- ||jpil-account pension scheme. New Zealand voters overwhelmingly rejected rrffhii pension restructuring in 1997, leaving New Zealand with its existing SsKystem of flat-rate universal benefits.

Sweden *

^Sweden currently is among the oldest countries in the world in terms of its ^^Hilation over age 65. Sweden's generous public pension system—until

 

 

the late 1990s a universal pension with earnings-related pension on top and an income-tested pension supplement for those with low-earnings histories—was very effective at poverty reduction, but also very expensive.

Swedish political institutions offer limited opportunities for imposing unpopular policy changes. The Social Democratic Party has been the dominant party in Sweden's single-chamber (since 1970) legislature over the past half-century, holding power for all but nine years. But a long decline in the Social Democrats' vote share, persistent minority governments, and short electoral cycles (three years between 1970 and 1994, after which the cycles returned to four years) all make loss imposition difficult.

Sweden experienced an extremely severe fiscal and economic crisis in the early 1990s, with both unemployment and budget deficits soaring to unprecedented levels. Because both payroll taxes and the overall tax burden were very high, increasing taxes was not seen as a viable option. Instead, Sweden turned to multiple rounds of incremental retrenchment in pension policy. At the same time; however, the nonsocialist government elected in Sweden in 1991 began a multiparty negotiating process for a comprehensive reform that was continued by the Social Democrats when they returned to power in 1994. This process was notable in several respects: It included most of the "coalitionable" (i.e., likely to serve in government) parties on both sides of the left/right divide; it was dominated by pragmatists on both sides; employer and union "social partners" were excluded from direct participation; and participating parties agreed to stick to the principles of agreement. Use of a multiparty working group helped to shield any particular party or set of parties from blame. The Riksdag endorsed the working-group principles in the summer of 1994, but a final agreement was not enacted until 1998, in large part because of opposition at the grassroots of the Social Democratic Party and the blue- collar labor union confederation.²⁴

The new Swedish pension system is a major change in concept from the old one, although its impact is being phased in gradually. The universal pension tier is being phased out, and a new earnings-related pension will be less generous than its predecessor. The new pension system is based on what Swedes call "notional defined contribution" principles: Benefits are paid out based on the entire lifetime of contributions, with contribution rates stabilized permanendy. The new system also includes a parallel system of individual accounts. The risks of poor economic performance and increased longevity have been shifted from the state to workers through a system of "automatic balancing mechanisms" that will insulate politicians

 

 

income-tested pension.

Germany

Germany faces a very serious challenge in the near term because of the aging of its population and an even greater challenge in the longer term, exacerbated by very low fertility rates. Germany relies overwhelming on a single pay-as-you-go social-insurance pension tier with generous provision for early retirement. The task of financing public pensions was further exacerbated by the post-1989 decision to use the pension system to help finance the costs of German reunification by promoting early labor- market exit in the former East Germany. With pension payroll tax rates already approaching 20 percent by the late 1980s (with additional subsidies from general revenues), pressures for retrenchment were enormous.

Germany's MMP electoral system has resulted in consistent coalition governments in recent years in the Bundestag. But the nature of coalition politics has changed as the party system has grown more fragmented in recent years and a two-block (Christian Democrat/Christian Social Union/Liberal versus Social Democrats and Greens) party system has . emerged. The fact that the second chamber (Bundesrat) has veto power on . some issues and is sometimes controlled by opposition parties further "complicates pension reform efforts. Equally important, however, is the _; breakdown in the 1990s of a longstanding cross-party "pension consensus" -^mechanism that had incorporated employers and unions as well as the ; niajor political parties and helped to keep pension issues out of electoral 1 politics.

ffRising costs and payroll taxes have contributed to multiple rounds of pension retrenchment, including reductions in the generosity of early- vretirement benefits (although still less-than-complete actuarial reduc- ^tions), enactment (and later revocation when a new Social :.:{Democratic/Green coalition came to power) of a "demographic factor" Ethat would have automatically lowered benefits as life expectancy rises, additional reductions over time in replacement rates, and a closer linkage ; between contributions and benefits. Targets were also set for both near- ; term and longer-term caps on payroll tax rates.

gPb'Change has continued under the Social Democratic/Green coalition. £ After a prolonged debate, a new quasi-mandatory tax-advantaged individual account tier was enacted in 2001 to compensate for planned future

 

 

declines in public system replacement rates. And the coalition has re- enacted under a new label most of the features of the "demographic factor" it initially revoked after coming to power. Thus, while Germany has not moved as for in restructuring its pension system as Sweden, fundamental restructuring has begun and is likely to continue in the future.

Conclusion

The cases examined here suggest some mixed conclusions about the impact of governmental institutions on short-term loss-imposition and nonincremental policy restructuring. A number of common patterns in pension reform are visible across all types of governmental arrangements, notably a strong tendency for governments to use long lead times and "grandfathering" to avoid blame. Also striking is the frequency with which a variety of nonstandard legislative procedures was used to enact politically risky pension reforms. These include procedurally privileged budget procedures (Canada, United States, New Zealand) and legislation under Urgency rules (New Zealand). Nonstandard legislative procedures can also be seen in other countries' pension reforms, such as the use of executive decree authority in Italy. Overall, policy feedbacks appear to be more important and predictable determinants of outcomes than governmental institutions.

Some significant institution-specific patterns are also visible, however. With respect to the U.S. separation-of-powers system, multiple veto points clearly contributed to the extreme outlier status of the United States in not enacting any major policy change in Social Security since 1983. The structure of U.S. institutions can also help to explain why the pension reform agenda has remained relatively broad (agenda control is not monopolized by political executives), and why presidents and legislative leaders do not press very hard for restructuring reforms that are unlikely to get enacted, as with Social Security privatization proposals in the first term of George W. Bush.

The pension reform records of Westminster-style parliamentary systems, on the other hand, vary greatly, ranging from a major restructuring and cutbacks in the United Kingdom under Thatcher, to frequentJinker- ing in that country under Blair, to modest and intermittent change in Canada, to highly volatile short-term retrenchment as well as one major restructuring enacted and immediately reversed in New Zealand. The New

 

 

Zealand case in particular suggests that, even where short-term losses are imposed in Westminster systems, they may not be sustainable over time. The United Kingdom under Thatcher was an outlier, even among Westminster systems, in its capacity to impose both short-term losses and a shift in policy regimes. Westminster institutions were a necessary but insufficient cause: long electoral cycles, favorable policy inheritances, divisions among the political opposition, and a political leader willing to take risks were all important contributing causes to the Thatcher government's success at pension reform.

Proportional representation-based parliamentary systems also show very high variation on loss-imposing capacity. For example, Sweden was able to enact a major restructuring of its public pension system, whereas Germany enacted a less fundamental reform and New Zealand voters rejected a major reform proposal in a 1997 referendum. •^h The countries examined here also suggest that ad hoc mechanisms may act as functional substitutes for concentrating power and/or diffusing blame. In particular, there is a recurring pattern of blame-diffusing - arrangements between major political actors that is used in pension policy. These arrangements, which might be called "policy cartels," take different forms: a multiparty pension working group in Sweden, a bipartisan 2 Social Security Reform Commission in the United States from 1981 to 1983, £ multiparty pension consensus in Germany, a Multi-party Pensions Accord in New Zealand, and closed-door negotiations between federal and provincial finance ministers on Canada Pension Plan reform. Technocratic ffovernments in countries like Italy can also be considered to be a form of : multisector policy cartel, since they are tolerated by multiple political parities but insulate those parties from political blame for tough decisions gtaken.

|PWhy do governments use policy cartels and other ad hoc mechanisms ^relatively frequently rather than simply change the structure of governmental institutions? One obvious reason is that governmental institutions hard to change: They are frequently embedded in constitutions and/or San be changed only by the vote of a supermajority. Moreover, existing ifofmal institutions are also likely to be beneficial to participants in other Sectors or for other purposes. Moreover, change in governing institutions Sreiites unpredictable risks. Policy cartels can serve as a sector-specific _ Adaptation to shortcomings in overall governmental institutions that p-adhieve acceptable outcomes without the costs and uncertainty of broader PSvernmental reforms.

 

 

The cases considered here also suggest, however, that there are significant cross-national differences in attempts at and success in cartelization. Policymaking cartels in pension policy have been largely absent in the United Kingdom, rare and unsustained in the United States and New Zealand, sustained but eventually collapsing in Germany, sustained in Sweden, and intermittent but hard to classify (since it involved federal- provincial negotiations rather than negotiations with social actors) in Canada. These arrangements are inherently hard to predict in a general analysis of governmental institutions, since they are by nature ad hoc. But they can be understood as arising from the desire of policymakers to "venue shift" from governmental arrangements that (1) have high political costs and/or (2) are unlikely to produce acceptable outcomes in a high- profile policy sector.

The cases suggest several conditions that are likely to make a pension policy cartel more successful and sustainable. A first condition is the existence of established patterns of cross-party/bloc policy cooperation on multiple issues, which helps to establish trust and mutual dependence. This is probably more likely to occur in PR systems than in Westminster ones, but, as Tsebelis argues, the distribution of participants' preferences as well as structure of institutions is a critical determinant of policy change.²⁵ Second, policy cartels are more likely to be sustainable when action-forcing mechanisms threaten large and unpredictable political costs if there is no cooperation, as in the case of the 1981-83 Social Security reform commission in the United States. Third, policy cartels may succeed when an opposition party or parties thinks that it is likely to win the next election in any event and doesn't want to deal with the issue in government. Also helpful are the presence of a labor-oriented party that is perceived by other parties to have a high capacity to prevent grassroots labor mobilization to overturn any agreement and make it an election issue and the availability of credible sanctions to prevent defection from the cartel. A pension policy cartel is less likely to be sustainable and successful in promoting controversial policy changes when there are veto points that are not controlled by members of the policy cartel, when party discipline is weak, when electoral cycles are short, and when parties negotiating an accord have weak credibility with voters.

The effects of governmental institutions on capacity for policy change are,, in short, uneven and heavily contingent on micro-rules and interaction with other causal forces. Indeed, institutional effects are probably best conceptualized as interaction terms with those other forces (e. g., separa-

 

 

tion of powers and divided government, Westminster systems and short electoral cycles), rather than as separate variables. But this pattern is by no means unique to governmental institutions, and the conclusion that social scientists should draw is not that we should give up studying the impact of these institutions. It is rather that we should proceed cautiously both in , the research that we do and in the conclusions that we draw from our research. More specifically, in designing research on the effects of institutions, it is important to:

• Consider multiple alternative hypotheses and causal chains
• Test our hypotheses about impacts across countries, time periods, and policy sectors to try to increase their robustness, and to find out where they need to be qualified
• Look for institutional adaptations such as policy cartels that fell outside the main categorizations used in institutional analysis and for "venue-shifting" that counteracts "normal" institutional characteris- tics

p^l - . Generalize cautiously and sparingly Йр^ • Draw conditional conclusions.

llfiGlearly, these cautions apply not only to the study of governmental insti- Kfiitions but to the study of labor market institutions, political parties, and Iggjndeed any set of institutional arrangements.

NOTES

jf^'i. Kathleen Thelen and Sven Steinmo, "Historical Institutionalism in Compara- llgtiye Politics," pp. 1-31 in Sven Steinmo, Kathleen Thelen, and Frank Longstreth, ВЩ'вй Structuring Politics: Historical Institutionalism in Comparative Politics (Cam- ||lfbrrdge:~Cambridge University Press, 1992).

Douglass C. North, Institutions, Institutional Change, and Economic Perfor- ИЩийИсе (Cambridge: Cambridge University Press, 1990), 3. ИдВрШ' Ellen Immergut, "The Theoretical Core of the New Institutionalism," Politics Щ^гШ Society 26,1 (1998), 5-34, at p. 5,