PrefBlog

The Bank of Canada has released the Bank of Canada Review – Autumn 2012 with articles:

• Financial Transaction Taxes: International Experiences, Issues and Feasibility by Anna Pomeranets
• Access, Competition and Risk in Centrally Cleared Markets by Jean-Sébastien Fontaine, Héctor Pérez Saiz and Joshua Slive
• Monetary Policy and the Risk-Taking Channel: Insights from the Lending Behaviour of Banks by Teodora Paligorova and Jesus A. Sierra Jimenez
• The Changing Landscape for Retail Payments in Canada and the Implications for the Demand for Cash by Carlos Arango, Kim P. Huynh, Ben Fung and Gerald Stuber

The first article attracted some notice from the Globe and Mail, in pieces by David Parkinson and Kevin Carmichael. An earlier working paper by Ms. Pomeranets and Daniel G. Weaver which focussed on the historical experience in New York State was reviewed on PrefBlog. This paper was quoted in support of the conclusion:

On balance, the literature suggests that an FTT is unlikely to reduce volatility and may instead increase it, which is consistent with arguments made by opponents of the tax.

The current paper is introduced with:

ƒƒ

• The financial transaction tax (FTT) is a policy idea with a long history that, in the wake of the global financial crisis, has attracted renewed interest in some quarters.
• Historically, there have been two motivating factors for the introduction of the tax. The first is its potential to raise substantial revenues, and the second is its perceived potential to discourage speculative trading and reduce volatility.
• There is, however, little empirical evidence that an FTT reduces volatility. Numerous studies suggest that an FTT harms market quality and is associated with an increase in volatility and a decrease in both market liquidity and trading volume. When the cost of acquiring a security rises, its required rate of return and cost of capital also increase. As a result, an FTT may reduce the flow of profitable projects, decreasing levels of real production, expansion, capital investment and even employment.
• There are many unanswered questions regarding the design of FTTs and their ability to raise significant revenues.

The imposition of a 20bp transaction charge in France (which has resulted in a greater interest in derivatives such as Contracts For Difference) was discussed on PrefBlog on November 15.

The authors also see fit to highlight:

Umlauf (1993) examines how financial transaction taxes (FTT s) affect stock market behaviour in Sweden. In 1984, Sweden introduced a 1 per cent tax on equity transactions, which was doubled to 2 per cent in 1986. Umlauf studies the impact of these changes on volatility and finds that volatility did not decline following the increase to the 2 per cent tax rate, but equity prices, on average, did decline.

Furthermore, Umlauf concludes that 60 per cent of the trading volume of the 11 most actively traded Swedish share classes migrated to London to avoid the tax. After the migration, the volatilities of London-traded shares fell relative to their Stockholm-traded counterparts. As trading volumes fell in Stockholm, so did revenues from capital gains taxes, completely offsetting the 4 billion Swedish kronor that the tax had raised in 1988.

Pomeranets also points out:

Critics of the FTT argue that it reduces market liquidity by making each trade more costly, simply because it is a tax and also because market forces react to it by offering fewer and lower-quality trading opportunities. The cost impact is evident in the way the FTT widens the bid-ask spread. Bid-ask spreads compensate traders for three things—order-processing costs, inventory risk and information risk—often called the three components of the bid-ask spread. The FTT will increase the costs of these three components in the following ways:…

And finally, we get the the social function of markets – capital formation:

Another measure of market quality examined in the literature is the cost of capital. Amihud and Mendelson (1992) conclude that a 0.5 per cent FTT would lead to a 1.33 per cent increase in the cost of capital. This result is consistent with their previous work that finds a positive relationship between required rates of return and transaction costs (Amihud and Mendelson 1986). When the cost of acquiring a security increases, its required rate of return and cost of capital also increase. As a result, an FTT would increase the cost of capital, which could have several harmful consequences. It could reduce the flow of profitable projects, shrinking levels of real production, expansion, capital investment and even employment.

Ms. Pomeranets concludes:

This article examines the main arguments regarding the costs and benefits of FTTs and explores some of the significant practical issues surrounding the implementation of an FTT. Little evidence is found to suggest that an FTT would reduce speculative trading or volatility. In fact, several studies conclude that an FTT increases volatility and bid-ask spreads and decreases trading volume. Furthermore, a number of challenges associated with the design and effectiveness of an FTT could limit the revenues that FTTs are intended to raise. For these reasons, countries considering the imposition of FTTs should be aware of their negative consequences and the challenges involved in implementation.

The second article examines a hobby-horse of mine – central clearing for derivatives, a dangerous policy recklessly promoted by the political establishment both directly and through their mouthpiece, Lapdog Carney. The last BoC attempt at justification, in the June 2012 Financial System Review was discussed on PrefBlog.

In this go-round, the authors state:

ƒƒ

• Central counterparties manage and mitigate counterparty credit risk in order to make markets more resilient and reduce systemic risk. Better management of counterparty risk can also open up markets to new participants, which in turn should reduce concentration and increase competition. These benefits are maximized when access to central counterparties is available to a wide range of market participants.
• In an over-the-counter market, there is an important trade-off between competition and risk. Concentrated, less competitive markets are more profitable and thus participants are less likely to default. But a central counterparty that provides sufficient access can improve this trade-off, since the gains from diversification—which will become greater as participation grows—can simultaneously reduce risk and increase competition.
• Regulators have developed, and central counterparties are implementing, new standards for fair, open and risk-based access criteria. Such standards will, among other things, counter any incentives that might exist for members of a central counterparty to limit access in order to protect their market share.

In other words, a major goal of Central Clearing is to provide employment for regulators, who will make fair and open, and, it must be emphasized, entirely corruption-free decisions regarding which smaller and and less creditworthy firms will be admitted to the club.

The crux of the matter is this:

The improved management of counterparty credit risk at a CCP opens markets to greater participation, which can increase competition. In OTC markets that are cleared bilaterally, participants are directly exposed to the risk that their counterparties may default and therefore have an incentive to restrict trading to counterparties that are known to be creditworthy. When a CCP with strong risk controls takes on the management of credit risk, however, participants can feel more secure trading with others—even anonymously— since the CCP guarantees that the terms of the trade will be honoured.

In other words, when the Bank of Downtown Plonksville enters into a trade with central clearing, its counter-party will charge exactly the same risk premium as it charges to the Bank of Canada. Some people consider this to be an advantage of the new regime.

If direct access to a CCP was limited to the largest dealers, their systemic importance would increase, potentially exacerbating the “too-big-to-fail” problem and preventing the CCP from providing the full benefits of diversification. Limited access could also make mid-tier institutions more vulnerable in times of stress and slow the transition to central clearing (Slive, Wilkins and Witmer 2011).

I wonder if the CCP itself is “too-big-to-fail” ….

The authors emphasize the importance of regulators and their awesomely wise, highly informed decisions throughout the process.

The third article is introduced with:

ƒƒ

• The financial crisis of 2007–09 and the subsequent extended period of historically low real interest rates in a number of major advanced economies have revived the question of whether economic agents are willing to take on more risk when interest rates remain low for a prolonged time period.
• This type of induced behaviour—an increased appetite for risk that causes economic agents to search for investment assets and strategies that generate higher investment returns—has been called the risk-taking channel of monetary policy.
• Recent academic research on banks suggests that lending policies in times of low interest rates can be consistent with the existence of a risk taking channel of monetary policy in Europe, South America, the United
  States and Canada. Specifically, studies find that the terms of loans to risky borrowers become less stringent in periods of low interest rates. This risk-taking channel may amplify the effects of traditional transmission mechanisms, resulting in the creation of excessive credit.

This effect is also inherent in the offsetting behaviours of the “expectations” component and the “risk premium” component of long-term rates, discussed in the Summer 2012 review discussed in PrefBlog.

The results suggest that the difference in the all-in-drawn spreads between loans to risky and less-risky borrowers decreases when interest rates are low relative to periods when they are high. Accounting for loan, firm and bank balance-sheet factors, as well as yearly and quarterly factors, the results show that the difference in the all-in-drawn spread between risky and less-risky borrowers is 48 per cent smaller when interest rates are lower than when they are higher (based on the first definition). This result is also economically significant: it implies that the difference in loan rates between risky and less-risky borrowers is 107 basis points smaller when the rates are low than when they are high.

The fourth article, of great interest to those in the field and to the Bank, but of somewhat less importance to other investors, is summarized as:

ƒƒ

• The share of cash in overall retail payments has decreased continuously
  over the past 20 years.

• Recent Bank of Canada research on consumers’ choice of payment instruments indicates that cash is frequently used for transactions with low values because of its speed, ease of use and wide acceptance, while debit and credit cards are more commonly used for transactions with higher values because of perceived attributes such as safety and record keeping.
• While innovations in retail payments currently being introduced into the Canadian marketplace could lead to a further reduction in the use of cash over the longer term, the implications for the use of cash of some of the structural and regulatory developments under way are less clear.
• The Bank of Canada will continue to monitor various developments in retail payments and study their implications for the demand for cash over the longer term.

This post is really late, I know. But I’m catching up slowly!

The Bank of Canada has released the Bank of Canada Review: Summer 2012 with the articles:

• Measurement Bias in the Canadian Consumer Price Index: An Update
• Global Risk Premiums and the Transmission of Monetary Policy
• An Analysis of Indicators of Balance-Sheet Risks at Canadian Financial Institutions

The first article, by Patrick Sabourin, makes the point:

Commodity-substitution bias reflects the fact that, while the weights of items in the CPI basket are held constant for a period of time, a change in relative prices may cause patterns in consumer spending to change. If, for example, the price of chicken were to increase considerably following supply constraints, consumers would likely purchase less chicken and increase their consumption of beef, since the two meats may be perceived as substitutes for each other. The CPI, however, assumes that consumers would continue to purchase the same quantity of chicken following a price change. This means that the measured change in the CPI will overstate the increase in the minimum cost of reaching a given standard of living (i.e., there is a positive bias).

I’ve always had trouble with this concept. I love beef. I despise chicken. As far as I am concerned, there is a separate quality adjustment that must be made that would mitigate, if not completely offset, the substitution adjustment when beef becomes too expensive and I have to eat chicken.

And, I am sure, this occurs for every other possible substitution. Although I might try explaining to my girlfriend that Coach handbags have become too expensive and I will follow theoretically approved procedure and get her, say, a plastic shopping bag for Christmas instead.

The second article, by Gregory H. Bauer and Antonio Diez de los Rios examines the relationship between long- and short-term interest rates:

ƒƒ

• An important channel in the transmission of monetary policy is the relationship between the short-term policy rate and long-term interest rates.
• Using a new term-structure model, we show that the variation in long-term interest rates over time consists of two components: one representing investor expectations of future policy rates, and another reflecting a term-structure risk premium that compensates investors for
  holding a risky asset.

• The time variation in the term-structure risk premium is countercyclical and largely determined by global macroeconomic conditions. As a result, long-term rates are pushed up during recessions and down during times of expansion. This is an important phenomenon that central banks need to take into account when using short-term rates as a policy tool.
• We illustrate this phenomenon by showing that the “conundrum” observed in the behaviour of long-term interest rates when U.S. monetary policy was tightened during the 2004–05 period was actually part of a global phenomenon.

In their model:

The long-term rate is decomposed into two terms in the following equation:
…
The first term involves market expectations, that is, the average expected 1-year interest rate over the next 10 years. In our model, we use the 1-year interest rate in country j as a proxy for that country’s policy rate. Observed yields will, on average, equal the expectations component only under the “expectations hypothesis,” which has been statistically rejected in many studies.

The rejection of the expectations hypothesis is typically attributed to the existence of the second term in equation (1), a time-varying term-structure risk premium. The risk premium represents the extra compensation that investors require for holding a 10-year bond. In our model, agents hold portfolios for one year, and the prices of long-term bonds may change considerably over that period, necessitating a higher expected rate of return. Several studies have focused on the properties of the term-structure risk premium (see Cochrane and Piazzesi (2005) and their references).
…
The second real-world aspect of the model consists of the constraints placed
on the time-varying risk premium, the second component of equation (1). Previous work has shown that imposing restrictions on the term-structure risk premium makes the forecast values of interest rates more realistic than those in unrestricted models.7 Our model restricts risk premiums on bonds through its assumption of global asset pricing; i.e., in integrated international markets, only global risks carry significant risk premiums. As a result, the term-structure risk premium on any bond is driven by the bond’s exposure to the global level and slope factors only. The local factors, while helping to explain prices at a point in time, do not affect expected returns (i.e., changes in prices), since investors can eliminate their effects by diversifying with a global portfolio.

Click for Big

Click for Big

This effect is evident during the financial crisis of 2007–09. While short-term U.S. rates fell by 263 basis points, long-term U.S. rates decreased by a mere 23 basis points. This occurred because, although the Fed succeeded in lowering expectations of future policy moves by 224 basis points (Table 2), the term-structure risk premium rose by 190 basis points.
…
The analysis in this article demonstrates the extent to which the global term-structure risk premium as well as monetary policy actions influence long-term interest rates. The risk premium is countercyclical to the global business cycle and thus may affect long-term interest rates in the opposite direction to that related to central bank policy actions. As a result, central banks need to take these forces into account in appropriately calibrating their policy response. Indeed, given the current low level of long-term rates, understanding movements in the global risk premium is important for the monetary policy decision-making process.

Since monetary policy may affect expectations and the term-structure risk premium differently, the levels of these two components may, in turn, affect the macroeconomy in various ways. For these reasons, understanding the effects on growth and inflation of movements in market expectations and the global term-structure risk premium is an important aim for future research.

The third article, by David Xiao Chen, H. Evren Damar, Hani Soubra and Yaz Terajima, will be of interest to students of Canadian banking and regulation thereof:

ƒƒ
• This article compares different types of Canadian financial institutions by examining over time ratios that are indicators of four balance-sheet risks—leverage, capital, asset liquidity and funding.
• The various risk indicators have decreased during the past three decades for most of the non-Big Six financial institutions in our sample and have remained relatively unchanged for the Big Six banks, resulting in increasing heterogeneity in these indicators of balance-sheet risks.
• The observed overall decline and increased heterogeneity in the risk indicators follow certain regulatory changes, such as the introduction of liquidity guidelines on funding in 1995 and the implementation of bank specific leverage requirements in 2000. This suggests that regulatory changes have had significant and heterogeneous effects on the management of balance sheets by financial institutions and, given that these regulations required more balance-sheet risk management, they contributed to the increased resilience of the banking sector.

Click for Big

Of particular interest is the funding ratio, defined as:

we define a funding ratio as the proportion of a bank’s total assets that are funded by wholesale funding (a relatively less stable funding source than retail (personal) deposits, for example):

Funding ratio (%) = 100 x (Non-personal deposits + repos)/Total assets.

A higher funding ratio indicates that a bank relies on greater market-based funding and is therefore more exposed to adverse shocks in the market that could disrupt continuous funding of its assets.

The Funding Ratio is of great interest due to Moody’s recent highlighting of:

the large Canadian banks’ noteworthy reliance on confidence-sensitive wholesale funding, which is obscured by limited public disclosure, increases their vulnerability to financial markets turmoil.

The BoC paper also highlights the eagerness of the politicians to inflate the housing bubble:

In addition, the growing popularity of mortgage-loan securitization in the late 1990s, following the introduction of the Canada Mortgage Bonds Program, raised the percentage of mortgage loans on bank balance sheets, especially among large and medium-sized financial institutions.(note)

Note reads: Increasing demand for mortgage loans caused by demographic shifts and lower down-payment
requirements has also played a role. See Chen et al. (forthcoming) for more details.

The authors conclude:

This article analyzes the balance-sheet ratios of Canadian financial institutions. Overall, various measures of risk have decreased over the past three decades for most non-Big Six institutions and have remained relatively unchanged for the Big Six banks. We find that smaller institutions, particularly trust and loan companies, generally have lower leverage and higher capital ratios than other types of financial institutions, including the Big Six banks. They also have larger holdings of liquid assets and face lower funding risk compared with other financial institutions. The observed overall decline and increased heterogeneity in risk (as measured by divergent trends in the leverage, capital and asset-liquidity ratios) followed certain regulatory changes, such as the introduction of liquidity guidelines on funding in 1995 (which preceded a sharp decline in, and more dispersion of, funding ratios among non-Big Six institutions) and the implementation of bankspecific leverage requirements in 2000 (which preceded a divergence in leverage ratios between the Big Six and non-Big Six institutions). This suggests that regulatory changes had significant and heterogeneous impacts on the management of balance sheets by financial institutions, resulting in the increased resilience of the banking system. While market discipline may have also played a role, more research is needed to identify changes in the degree of market discipline in the Canadian banking sector.

Given the observed variation in behaviour among Canadian financial institutions, continued analysis of different types of institutions can enable a more comprehensive assessment of financial stability. Understanding the different risks faced by various types of financial institutions improves the framework that the Bank of Canada uses to monitor developments of potential risks in the banking sector.

The statement that “This suggests that regulatory changes had significant and heterogeneous impacts on the management of balance sheets by financial institutions, resulting in the increased resilience of the banking system.” strikes me as being a little bit fishy. Regulatory change did indeed have “significant and heterogeneous impacts on the management of balance sheets by financial institutions”, but whether this resulted “in the increased resilience of the banking system.” has not been addressed in the paper. That was the intention, certainly, and may well be true, but a cause and effect relationship has not been demonstrated.