Central bank independence: from theory to practice
Speech by Lorenzo Bini Smaghi, Member of the Executive Board of the ECB*
at the conference
Good Governance and Effective Partnership
Budapest, Hungarian National Assembly, 19 April 2007
The issue of central bank independence has been the subject of important academic work. However, the literature has mainly focussed on the theoretical and formal aspects, without considering that the distance between theory and practice is not always so short.  The experience of the past few years has shown that the implementation of the rule of law is not without challenges, even inside the European Union.
The basic point I would like to make today is that the legal provisions are necessary, but not sufficient to ensure central bank independence. The incentives to circumvent the legal framework, with a view to influencing the behaviour of the central bank, are always looming. Therefore, “strong vigilance” - to use words that have a clear meaning in the world of central banks - is required to ensure that these incentives do not translate into actions. This is not only a task for the central bank or for political authorities, at the national and supranational levels, but for society as a whole, including the media and public opinion.
I will not dwell too much on the benefits of central bank independence. These have been extensively examined and are generally undisputed, at least at an academic level. It is easy to understand why central bank independence is essential in modern monetary systems. In a paper-money regime, where government liabilities represent means of payment and have purchasing power, there is always the temptation for any government to use such money in an opportunistic manner. The temptation comes from the fact that money creation has positive effects in the short term, on growth and employment, while the costs, in terms of higher inflation, are paid over the medium to longer term.  Central bank independence is a way to protect policy makers against the temptation of using monetary policy in a distortionary way. However, the temptation is not eliminated. This might be a reason why central bank independence appears to be strongly supported by the citizens. 
There is ample empirical evidence that central bank independence brings about lower inflation, which ensures a more stable environment for economic and employment growth. 
In the European Union, the principle of central bank independence has a quasi-constitutional basis. Article 108 of the Treaty establishing the European Community states that:
“ neither the ECB, nor a national central bank … shall seek or take instructions from Community institutions or bodies, from any government of a Member State or from any other body”.
Article 7 of the Statute of the ESCB echoes this statement. The Treaty provisions on central bank independence apply to all EU Member States (except the United Kingdom, which has a special derogation ), irrespective of euro area membership. Countries are thus expected to have completed the process of granting their central bank full institutional independence by the time of accession to the EU; in practice, however, this is not the case.
This is a particular concern because it suggests that in several EU member states there is not yet a full understanding, in particular by the political authorities and possibly also by society as a whole, of the reasons underlying central bank independence and of the commitments that these countries have made to grant such independence at the time of accession. Central bank independence is instead often considered only as a pre-condition for adopting the euro. Failure to fully ensure central bank independence is, in fact, a breach of the Treaty which would justify bringing an action before the European Court of Justice. The European Commission, which has the power to take such an initiative according to the Treaty, should consider doing so when needed.
The Treaty provisions are in fact very clear, in specifying that central bank independence entails obligations for the political authorities. In particular, Article 108 states that:
“Community institutions and bodies and the governments of the Member States undertake to respect this principle [of central bank independence] and not to seek to influence the members of the decision-making bodies of the ECB or of the national central banks in the performance of their tasks”
The extent and nature of central bank independence can be assessed on the basis of its legal provisions. However, central bank independence also hinges on a broad series of factors and customary practices, which are partly determined by historical developments in the different countries. In particular, the way in which certain conflicts with other bodies of government have been resolved influences the extent to which a central bank is effectively protected against external interferences and marks the boundaries of independence. 
In November 1995 the European Monetary Institute (the precursor of the ECB) published a list of requirements for central bank independence. Over the years, the concept has been developed in the convergence reports as well as in the ECB opinions on draft national legal instruments referring to central bank law. 
Four categories of central bank independence can be considered: functional, institutional, personal and financial. Institutional independence is largely treated in the ECB convergence reports. Today, I will mainly discuss the other three aspects of independence. In addition, I will also touch upon three other issues that are related to or may affect central bank independence:
central bank involvement in prudential supervision;
the nexus between independence and accountability;
the cooperation and dialogue with other economic policy-makers.
I. Functional independence
An independent central bank should be free to set its policy instrument with the aim of achieving its objective. Functional independence thus requires that the primary objective of the national central bank of an EU member state be set in a clear and legally certain way and be fully in line with the primary objective of price stability established by the Treaty.
From an operational viewpoint, this implies that the central bank should have full autonomous power in setting the level of the short-term interest rate in the money market. In market economies , with modern financial systems, monetary policy normally operates through changes in the level of the interest rate. Any obstacle to the ability of central banks to affect market interest rates should be considered as an obstacle to their independence. An example of such an obstacle would be the obligation for a central bank to directly finance budget deficits, which would clearly reduce the ability to influence money market conditions in the direction it deems most appropriate for the pursuit of price stability. This is the reason why the Treaty establishes, in Article 101, that monetary financing of budget deficits is prohibited. This is, obviously, also a legal precondition for joining the EU and the euro and is regularly assessed in the ECB’s convergence reports. Other large industrial countries have adopted similar legislation over the years. 
More generally, governments can undermine the independence of monetary policy by conducting an excessively profligate and hence unsustainable fiscal policy. Although there have been many academic debates on this issue,  it is beyond doubt that conducting an independent monetary policy, aimed at the achievement of low and stable inflation, is made significantly more difficult by the existence of large budget deficits. This is true for two related reasons. First, when deficits and public debt become unsustainable, the incentive for the government to force the central bank to monetise its deficit, thus eliminating public debt via inflation, increases substantially. Second, the larger the budget deficit and the accumulated debt, the more market participants become aware of the risk of monetisation. In addition, they may believe that the central bank will be forced to “bail out” the government by assuming its liabilities, even if Article 103 of the Treaty explicitly prohibits this. This may jeopardise the anchoring of inflation expectations and make the control of inflation more costly. In this case, fiscal policy may become dominant over monetary policy, thus undermining, de facto if not de jure, the functional independence of the central bank. This relationship is admittedly more complex and less direct in a supranational environment like the euro area, with decentralised fiscal policies and a centralised monetary policy, but the basic nexus is the same.
Therefore, sound public finances, which in any event represent a valid objective per se, are an important underpinning of central bank independence. The way to achieve budgetary discipline with a view to protecting central bank independence may vary across countries. In the EU, explicit provisions on budgetary surveillance are foreseen in the Treaty and in the Stability and Growth Pact.
II. Personal independence
Turning to personal independence, the nomination and dismissal of the Governor and of the members of the central bank’s decision-making bodies pertain to the political authorities. This does not mean, however, that there should be no criteria. I would like to raise four issues in this regard: the term of office; professional qualifications; political affiliations; and collegiality.
Let me begin with the term of office. Article 14.2 of the Statute of the ESCB provides that the central bank laws must provide for a minimum term of office of five years for the Governor and protects against the arbitrary dismissal of Governors. Governors may be relieved of their duties only if they no longer fulfil the conditions required for the performance of their duties or if they have been guilty of serious misconduct. The possibility of recourse to the European Court of Justice is foreseen against any abuse of this provision.
One possibility to circumvent this provision can take place on the occasion of the adoption of new legislation aimed at reorganising the term of office of the Governor. The change in the law could be used as an opportunity to dismiss the Governor. In order to protect the personal independence of central bank Governors, any reorganisation of the term of office should foresee that the existing Governor can continue to perform his duties until the end of the term of office he has been appointed for. This was made clear, for example, in an ECB opinion on an Italian draft law changing the Governor’s term of office to eight years. While the term of office was compatible with the Treaty and the Statute, the ECB pointed out that the wording of the draft law amounted to an ex lege revocation of the appointment of the incumbent Governor.  The law was adapted accordingly.
Another possibility to circumvent the requirement of personal independence is to adopt different rules for the other members of the decision-making bodies of the national central banks involved in the performance of ESCB-related tasks. Especially in collegiate decision making bodies, if the other members do not have the same degree of independence as the Governor, the personal independence of the central bank is at risk. The ECB has made this point clear in a number of legal opinions over the past few years.  For instance, the ECB has advised the French  and Cypriot  legislators to amend provisions in national law that enabled the appointment of board members for a term of less than five years. As another example, a draft Slovak law foresaw alternative grounds for dismissal for the board members other than the Governor that were incompatible with Article 14.2 of the Statute. The ECB was of the opinion that such a formulation stood in the way of legal clarity and, more importantly, could be used to circumvent the effective protection of personal independence. The ECB therefore recommended referring only to Article 14.2 of the Statute of the ESCB when stating the reasons for dismissal of a board member.  To the ECB’s knowledge, this opinion has not yet been incorporated in central bank law.
Let me turn now to the issue of the professional qualification of the persons nominated to be members of the decision-making bodies. This is not only a substantive requirement but also a formal one. Indeed, it is not sufficient that the person has the professional qualifications, he/she must also be perceived to have them by the public. This would ensure that personal issues cannot be used to put pressure on the decision making bodies to influence their behaviour.
An important requirement for personal independence is the absence of any actual or potential conflict of interest between the duties related to the central bank decision-making bodies (and also in relation to the ECB for national central bank Governors) and any other functions which they may have. Hence, as a matter of principle, the membership of a decision-making body involved in the performance of ESCB-related tasks is incompatible with the exercise of other functions that might create a conflict of interest.
Furthermore, to effectively ensure central bank independence, the appointed members of central bank decision-making bodies should be clearly perceived to possess high professional capabilities and be super partes. The Statute of the ESCB states that:
“the President, the Vice-President and the other members of the [ECB] Executive Board shall be appointed from among persons of recognized standing and professional experience in monetary or banking matters”
Problems may arise if the members of the decision making bodies have a political affiliation or have played an active political role prior to their appointment, or (even more problematic) if they are expected to play a role afterwards. There are several examples showing that political affiliation can be used by other institutions or political parties as an excuse to put pressure on the central bank, especially after a change in government. The ability of the central bank to speak out, if needed, possibly in critical terms, with respect to economic and budgetary policies might also be impaired if the Governor, or the other members of the board, are perceived to have a political affiliation, as this might be interpreted as political interference. In this respect, the ECB has emphasised in its latest Convergence Report that the assessment of legal convergence should not be limited to a formal assessment of the letter of national legislation but may also consider whether the implementation of the relevant provisions complies with the spirit of the Treaty and the Statute. The ECB was particularly concerned about recent pressures being put on the decision-making bodies of some Member States’ national central banks, for instance in Poland last year, which were inconsistent with the spirit of the Treaty.
A final issue is related to collegiality. A collegial decision-making body, composed of several members is more likely to resist external pressures and partisan behaviour. The ECB has made this clear in some legal opinions, for example on the reform of the statute of the Banca d’Italia. 
III. Financial independence
A central bank cannot credibly operate in an independent way without proper financial means; it would be under a “Damocles’ sword” if it depends on the government for the financing of its operative expenses. The concept of financial independence should therefore be assessed from the perspective of whether any third party is able to exercise either direct or indirect influence not only over central bank tasks but also over its ability (understood both operationally, in terms of manpower, and financially, in terms of appropriate financial resources) to fulfil its mandate.
Four aspects of financial independence – the right to determine its own budget; the application of central bank-specific accounting rules; clear provisions on the distribution of profits; and clearly defined financial liability for supervisory authorities – are particularly relevant in this respect, and some of them have only been refined quite recently in some ECB legal opinions.  These are the features of financial independence where national central banks are most vulnerable to outside influence.
In the euro area and the European Union, Member States should not put their national central banks in a position where they have insufficient financial resources to carry out their ESCB - or Eurosystem - related tasks. Moreover, the principle of financial independence implies that a national central bank must have sufficient means not only to perform ESCB-related tasks but also its own national tasks (e.g. financing its administration and own operations).
This has not always been uncontroversial. For example, the Finnish legislator proposed in 2003 to reduce Suomen Pankki’s capital. The legislative proposal would have forced Suomen Pankki to sell part of its foreign reserve assets. The ECB indicated that reducing the equity of one national central bank and imposing restrictions on the management of its financial resources could not be considered in isolation from the potential effect of such a reduction on the financial position, and therefore on the financial independence, of the Eurosystem as a whole. The ECB noted that the obligation of Member States to ensure the independence of their central banks puts them in an exceptional position, since it obliges the Member States to keep the assessment of the level of financial resources and the management of the capital of the national central banks at arm’s length. The national central bank should not be dependent for its finances on the government, the parliament or any other third party. 
Another example of a similar kind can be taken from the discussion over the Central Bank and Financial Services Authority of Ireland Bill. The legislative proposal aimed to establish the regulatory authority as a constituent part of the central bank, with its management separated from that of the central bank. This structure could have posed risks for the national central bank’s financial integrity, since there was potential for the national central bank to be exposed to liabilities resulting from funding and budgetary decisions for the regulatory authority. The ECB therefore recommended giving the organs of the central bank controlling powers with regard to funding, budgetary and staffing decisions .
IV. Central bank independence and prudential supervision
The traditional borders between the banking, securities and insurance sectors of the financial market are becoming increasingly blurred, as demonstrated by the emergence of hybrid financial products, the increased use of risk transfer instruments and distribution agreements between the three sectors, and the growing role of financial conglomerates. In this context, the ECB has welcomed institutional frameworks established in Member States that recognise the essential role of central banks in promoting the safety and soundness of financial institutions and the stability of the financial system as a whole. 
In a number of legal opinions, the ECB has emphasised that maintaining the close involvement of national central banks in prudential supervision is an important condition for allowing the Eurosystem to contribute adequately to monitoring the risks to financial stability in the euro area and to safeguard a smooth coordination between the central banking functions exercised at the supranational level and the supervisory functions carried out at the national level. 
The requirement of independence for prudential supervision has, however, been traditionally somewhat more controversial than in the case of monetary policy. On the one hand, this reflects the fact that prudential supervision is a matter where political interests are typically strong, for various reasons (for example, providing credit to an insolvent bank is ultimately financed out of taxpayers’ money). On the other hand, there is no clear provision for supervisory independence in Community legislation. Over recent years, however, a doctrine of supervisory independence has developed as a “best practice” and is now one of the Basel Core Principles of good supervision.
In this regard, one should distinguish between two cases: (i) where prudential supervision is a department of the central bank; and (ii) where supervision is a separate entity, though possibly connected in some way with the central bank.
Some Member States place their financial supervisory authorities within their national central bank. If the law provides for separate decision-making for such supervisory authorities, it is important to ensure that decisions adopted by the latter do not endanger the finances of the central bank as a whole. In such cases, national legislation should enable the national central banks to have ultimate control over any decision taken by the supervisory authorities that could affect central bank independence, in particular its financial independence.  For example, under the Central Bank and Financial Services Authority of Ireland Act that I mentioned earlier, the regulatory authority has a duty to act in a way that is consistent with the performance by the Governor and the Board of their respective functions in relation to the Bank; for the purpose of verifying compliance with that duty, the regulatory authority must provide the Governor or the Board with the information that may be required.
More generally, prudential supervision, by its very nature, lends itself to being at the centre of political interests. When the central bank is involved in supervision, therefore, there is an increased risk of pressure on its financial independence and on the personal independence of the members of its decision-making bodies, as witnessed in some EU countries in recent years. Therefore, in such cases, central bank independence should be monitored particularly attentively.
V. Accountability and independence
It goes without saying that independence does not mean arbitrariness. Central bankers should never forget that they are ultimately accountable for their policies to the elected politicians and to the public at large (including future generations). In this respect, it is in my opinion misleading to think of proper accountability as a “counterweight” to independence. I rather prefer to think of accountability as the “other side of the coin” of independence and to see the two concepts as mutually reinforcing, rather than antagonists, as is sometimes suggested.
In order to be independent, central banks must have a clear objective against which their performance can be measured. The ECB’s mandate of maintaining price stability is a good example of a clear objective which fosters both independence and accountability. It is in fact easier for external parties to assess the performance of an independent institution if the criterion against which the performance is evaluated is clear and transparent. A complex and obscure objective, by contrast, naturally leads to increased margins for discretion by the independent authority and to a higher degree of opacity in its behaviour. Any weakening of the democratic control over an independent institution may lead to excessive discretion and unclear objectives, which risks creating political backlashes against independence and may over time undermine independence itself. Therefore, it is my firm view that independence is sustainable in the long term only if accompanied by strong accountability and transparency in the operations of the independent institution. The legal provisions can more easily be circumvented if there are no provisions ensuring transparency and accountability.
VI. Independence and dialogue with the government
As I come towards the end of my remarks, the last practical issue I would like to discuss today is the nexus between independence of different authorities on the one hand, and cooperation on the other. Economic policy-makers should have a strong interest in sharing information and maintaining a close dialogue, even in the presence of a clear allocation of tasks. Such is the case in the euro area, where there is a clear separation of responsibilities, with monetary policy being assigned to an independent central bank, and fiscal and structural policies assigned to national governments. This separation of responsibilities should not, and indeed does not, prevent a continuous and fruitful dialogue between the various institutions. The dialogue and cooperation between the ECB and the Eurogroup clearly does not jeopardise the independence of the former, nor the prerogatives of the latter.
In this respect, it is not clear what the calls for a stronger dialogue between the ECB and the Eurogroup, which are occasionally made in public, really aim at. Given that the framework for such a dialogue already exists, and can be fully exploited, one may suspect that behind such calls hides the desire for some political oversight over the ECB. In my opinion, this would be tantamount to undermining the ECB’s independence.
Incidentally, the experience of the US is often mentioned by those who call for a stronger coordination between the Eurogroup and the ECB, often forgetting that monetary policy is never discussed at meetings between the Fed Chairman and the Secretary of the Treasury, as confirmed in the memoires of former Treasury Secretary Robert Rubin. 
The calls on the ECB to be more cooperative, or even the attempts to influence its behaviour, do not seem to have affected the independence and credibility of the ECB, as is suggested by the lack of reaction of long-term inflation expectations as well as movements of the euro area yield curve. This is, on the one hand, reassuring, since it implies that the ECB’s independence rests on solid foundations and that markets do not pay much attention to the attempts, even by Finance Ministers, to influence monetary policy. On the other hand, it also shows that a culture of respect for central bank independence is still not fully established in Europe, not even in the euro area. This is not a good example for the other countries, including the new EU Member States which have to abide by the same provisions for central bank independence as in the euro area.
In conclusion, I would like to summarise my main argument: safeguarding central bank independence takes more than a series of legal provisions, as the experience of the past eight years has shown. It requires, above all, a wide degree of acceptance of the principle of independence within the underlying political and economic culture of the society. This requires a leading role by governments and politicians, especially those in the euro area but also in the other EU countries, in understanding and explaining the fundamental reasons behind the choice to delegate powers to an independent monetary authority for the welfare of present and future generations. Central bank independence, like any other law, needs to be continuously protected and implemented over time. This is the responsibility of the political institutions.
Thank you for your attention.
* The opinions expressed are those of the author. I thank A. Tupits and L. Stracca for their input in the preparation of this speech.
 Some academics have even proposed quantitative indicators to measure central bank independence; see for example Grilli , Masciandaro and Tabellini (1991): “Political and Monetary Institutions and Public Financial Policies in the Industrial Countries”, Economic Policy, 13, 341-92; and Cukierman, Webb and Neyapti (1992): “Measuring the Independence of Central Banks and its Effect on Policy Outcomes”, World Bank Economic Review 6(3) : 358-98.
 See L. Bini Smaghi (2006): Democratic Representation and Economic Policy Rules, Eunomiamaster, 28 September.
 According to the latest ECB opinion survey in the euro area countries excluding Slovenia, a large majority of citizens support the ECB’s independence from governments.
 For example, a recent paper by Cecchetti et al (2007) documents the large fall in inflation volatility after the mid-1980s, a time when monetary policy started to be run in an independent manner and with a clearer focus on the primary objective of price stability. See Cecchetti, Hooper, Kasman, Schoenholtz, and Watson (2007): "Understanding the Evolving Inflation Process," presentation at the U.S. Monetary Policy Forum, Washington, D.C., March 9.
 The Bank of England, in any event, is also largely independent.
 See L. Bini Smaghi and D. Gros (2000): Open Issues in European Central Banking, MacMillan, London.
 The ECB must be consulted by the Community institutions and the Member States on draft legislative provisions in its fields of competence, pursuant to Article 105(4) of the Treaty establishing the European Community and Article 4 of the Statute. Council Decision 98/415/EC of 29 June 1998 on the consultation of the ECB by national authorities regarding draft legislative provisions (OJ L 189, 3.7.1998, p. 42) expressly requires that the Member States take the measures necessary to ensure compliance with this obligation.
 Where the government has no power to set (most) prices directly.
 The Federal Reserve System of the United States has a prohibition to directly underwrite government debt in the Federal Reserve Act (FRA). This prohibition was not part of the original FRA of 1913, but was added after the First World War. In Japan, the Public Finance Law prohibits the Bank of Japan in principle from underwriting government bonds and making loans to the government. However, this does not apply to “special cases” within an amount determined by the Diet.
 See, for example, W. Buiter (2002): “The fiscal theory of the price level: a critique”, Economic Journal, vol. 112, pp. 459-480.
 See paragraph 8 of ECB opinion CON/2004/16 of 11 May 2004 at the request of the Italian Ministry of Economic Affairs and Finance on a draft law on the protection of savings.
 See paragraph 8 of ECB opinion CON/2004/35 of 4 November 2004 at the request of the Hungarian Ministry of Finance on a draft law amending the Law on Magyar Nemzeti Bank; paragraph 8 of ECB opinion CON/2005/26 of 4 August 2005 at the request of Národná banka Slovenska on a draft law amending Act No 566/1992 Coll. on Národná banka Slovenska, as amended, and on amendments to certain laws; paragraph 3.3 of ECB opinion CON/2006/44 of 25 August 2006 at the request of the Banca d’Italia on the amended Statute of the Banca d’Italia; paragraph 2.6 of ECB opinion CON/2006/32 of 22 June 2006 at the request of the French Senate on a draft law on the Banque de France; and paragraphs 2.3 and 2.4 of ECB opinion CON/2007/6 of 7 March 2007 at the request of the German Ministry of Finance on a draft Eighth Law amending the Law on the Deutsche Bundesbank.
 See paragraph 2.6 of ECB opinion CON/2006/32 of 22 June 2006 at the request of the French Senate on a draft law on the Banque de France.
 The amending law was adopted on 15 March 2007.
 See paragraph 8 of ECB opinion CON/2005/26 of 4 August 2005 at the request of Národná banka Slovenska on a draft law amending Act No 566/1992 Coll. on Národná banka Slovenska, as amended, and on amendments to certain laws
 See ECB opinion CON/2006/44 of 25 August 2006 at the request of the Banca d’Italia on the amended Statute of the Banca d’Italia.
 The main formative ECB opinions in this area are the following: CON/2002/16 of 5 June 2002 at the request of the Irish Department of Finance on a draft Central Bank and Financial Services Authority of Ireland Bill, 2002; CON/2003/22 of 15 October 2003 at the request of the Finnish Ministry of Finance on a draft government proposal to amend the Suomen Pankki Act and other related acts; CON/2003/27 of 2 December 2003 at the request of the Austrian Federal Ministry of Finance on a draft Federal law on the National Foundation for Research, Technology and Development; CON/2004/1 of 20 January 2004 at the request of the Economic Committee of the Finnish Parliament on a draft government proposal to amend the Suomen Pankki Act and other related acts; CON/2006/38 of 25 July 2006 at the request of the Bank of Greece on a draft provision on the Bank of Greece’s powers in the field of consumer protection; CON/2006/47 of 13 September 2006 at the request of the Czech Ministry of Industry and Trade on an amendment to the Law on Česká národní banka..
 See paragraph 8 of ECB opinion CON/2003/22 of 15 October 2003 at the request of the Finnish Ministry of Finance on a draft government proposal to amend the Suomen Pankki Act and other related acts.
 See paragraph 6 of ECB opinion CON/2002/16 of 5 June 2002 at the request of the Irish Department of Finance on a draft Central Bank and Financial Services Authority of Ireland Bill.
 See, for example, paragraph 5 of ECB opinion CON/2004/16 of 11 May 2004 at the request of the Italian Ministry of Economic Affairs and Finance on a draft law on the protection of savings.
 See paragraph 4 of ECB opinion CON/2001/10 of 25 May 2001 at the request of the Austrian Ministry of Finance on a draft Federal law establishing and organising the financial market supervisory authority; paragraph 5 of ECB opinion CON/2001/35 of 8 November 2001 at the request of the German Ministry of Finance on a draft law establishing an integrated financial services supervision. See also Chapter 2 of ECB opinion CON/2006/15 of 9 March 2006 at the request of the Polish Minister of Finance on a draft law on the supervision of financial institutions, as well as paragraph 2.8 of ECB opinion CON/2006/53 of 16 November 2006 at the request of the Polish Minister for Finance on a draft law amending the Law on trading in financial instruments.
 CON/2002/16 of 5 June 2002 at the request of the Irish Department of Finance on a draft Central Bank and Financial Services Authority of Ireland Bill, 2002.
 R. Rubin and J. Weisberg, Tough choices: from Wall Street to Washington.