PrefBlog

The BoC has released a Staff Working Paper by Carola Conces Binder, Rodrigo Sekkel titled Central Bank Forecasting: A Survey:

Central banks’ forecasts are important monetary policy inputs and tools for central bank communication. We survey the literature on forecasting at the Federal Reserve, European Central Bank, Bank of England and Bank of Canada, focusing especially on recent developments. After describing these central banks’ forecasting frameworks, we discuss the literature on central bank forecast evaluation and new tests of unbiasedness and efficiency. We also discuss evidence of central banks’ informational advantage over private sector forecasters—which appears to have weakened over time—and how central bank forecasts may affect private sector expectations even in the absence of an informational advantage. We discuss how the Great Recession led central banks to evaluate their forecasting frameworks and how the COVID-19 pandemic has further challenged central bank forecasting. Finally, we consider directions for future research.
…
A seminal paper by Romer and Romer (2000) shows that Federal Reserve staff forecasts of inflation in the Greenbook are more accurate than private sector forecasts from Blue Chip, Data Resources, Inc., and the Survey of Professional Forecasters (SPF), and that access to Greenbook forecasts could have helped commercial forecasters improve forecasts. Indeed, “someone with access to both the Federal Reserve and commercial forecasts should not just put positive weight on the Federal Reserve forecast, but put little weight on the commercial one” (Romer and Romer, 2000, p. 438). Correspondingly, the mean squared errors of the Greenbook forecasts are around 25% lower than those of commercial forecasters at most horizons. For real GNP growth, the Fed’s informational advantage is most prominent at short horizons but varies at longer horizons.

To demonstrate the implications of the informational advantage they document, Romer and Romer (2000) show that the Fed’s monetary policy actions reveal some of their private information about the economic outlook. Thus, when the Fed raises the federal funds rate, forecasters revise their inflation expectations upward, which can help explain why interest rates at long horizons respond to monetary policy. Subsequent research has continued to probe the informational advantage of central banks and to examine how central bank expectations influence private sector expectations, even in the absence of a clear informational advantage.

And Central Bankers are examined again in an IMF article by RAGHURAM RAJAN titled More focused, less interventionist central banks would likely deliver better outcomes:

Central bankers of industrialized countries have fallen tremendously in the public’s estimation. Not long ago they were heroes, supporting feeble growth with unconventional monetary policies, promoting the hiring of minorities by allowing the labor market to run a little hot, and even trying to hold back climate change, all the while berating paralyzed legislatures for not doing more. Now they stand accused of botching their most basic task, keeping inflation low and stable. Politicians, sniffing blood and mistrustful of unelected power, want to reexamine central bank mandates.

Did central banks get it all wrong? If so, what should they do?
…
Yet stopping the postmortem at this point is probably overly generous to central banks. After all, their past actions reduced their room to maneuver, and not only for the reasons just outlined. Take the emergence of both fiscal dominance (whereby the central bank acts to accommodate the government’s fiscal spending) and financial dominance (whereby the central bank acquiesces to the imperatives of the market). They clearly are not unrelated to central bank actions of the past few years.

Long periods of low interest rates and high liquidity prompt an increase in asset prices and associated leveraging. And both the government and the private sector leveraged up. Of course, the pandemic and Putin’s war pushed up government spending. But so did ultralow long-term interest rates and a bond market anesthetized by central bank actions such as quantitative easing. Indeed, there was a case for targeted government spending financed by issuing long-term debt. Yet sensible economists making the case for spending did not caveat their recommendations enough, and fractured politics ensured that the only spending that could be legislated had something for everyone. Politicians, as always, drew on unsound but convenient theories (think modern monetary theory) that gave them license for unbridled spending.

Central banks compounded the problem by buying government debt financed by overnight reserves, thus shortening the maturity of the financing of the government and central bank’s consolidated balance sheets. This means that as interest rates rise, government finances—especially for slow-growing countries with significant debt—are likely to become more problematic. Fiscal considerations already weigh on the policies of some central banks—for instance, the European Central Bank worries about the effect of its monetary actions on “fragmentation,” the yields of fiscally weaker countries’ debt blowing out relative to those of stronger countries. At the very least, perhaps central banks should have recognized the changing nature of politics that made unbridled spending more likely in response to shocks, even if they did not anticipate the shocks. This may have made them more concerned about suppressing long rates and espousing low-for-long policy rates.

The Office of the Comptroller of the Currency released the OCC Mortgage Metrics Report for 22Q4:

• • As of December 31, 2022, the reporting banks serviced approximately 12 million first-lien residential mortgage loans with $2.7 trillion in unpaid principal balances (see figures 1 and 2). This $2.7 trillion was 22 percent of all residential mortgage debt outstanding in the United States.2
• • Overall mortgage performance this quarter improved from the fourth quarter of 2021. The percentage of mortgages that were current and performing at the end of the fourth quarter of 2022 was 97.1 percent compared with 96.4 percent at the end of the fourth quarter of 2021 (see figure 6). The CARES Act, signed into law on March 27, 2020, and extended on February 18, 2022, allows for loan forbearance that can extend up to 360 days and is reflected in the mortgage performance data.
• • Servicers initiated 9,166 new foreclosures in the fourth quarter of 2022, a decrease from the prior quarter, but a higher volume than a year earlier (see figure 7). The new foreclosure volume in the fourth quarter of 2022 is lower than pre-COVID-19 pandemic foreclosure volumes. Home forfeiture actions during the fourth quarter of 2022—completed foreclosure sales, short sales, and deed-in-lieu-of-foreclosure actions—increased 42.8 percent from a year earlier to 2,525 (see figure 8). Events associated with the COVID-19 pandemic, including foreclosure moratoriums that began March 18, 2020, and were extended to July 31, 2021, have significantly affected these metrics.

It doesn’t look like we’re in a recession. Not yet, anyway!

The Canadian economy has picked up momentum in the early stages of 2023 and avoided slipping into a recession, despite the highest interest rates in more than 15 years.

Real gross domestic product rose 0.5 per cent in January from the previous month, Statistics Canada reported on Friday. In a preliminary estimate, the agency said the economy grew by a further 0.3 per cent in February.

On an annualized basis, growth is trending toward 2.5 per cent in the first quarter – far stronger than the 0.5 per cent pace that the Bank of Canada had projected. This also marked a rebound from no growth in the fourth quarter, a result that was largely owing to a sharp pullback in inventory investments.

The BoC has released a Staff Working Paper by Serdar Kabaca and Kerem Tuzcuoglu titled Supply Drivers of the US Inflation Since the Pandemic:

This paper examines the contribution of several supply factors to US headline inflation since the start of the COVID-19 pandemic. We identify six supply shocks using a structural VAR model: labor supply, labor productivity, global supply chain, oil price, price mark-up and wage mark-up shocks. Our shock identification relies mainly on sign restrictions. But for the global supply chain shock, we propose a new identification scheme combining sign, narrative and variance decomposition restrictions. Historical decomposition results suggest that global supply chain and oil price shocks are the biggest supply contributors to the US inflation during the pandemic. In contrast, labor shortages only mildly contribute to inflation, but their impact on output is larger in that period. Additionally, price and wage mark-up shocks start to significantly contribute to inflation only towards the middle of 2022. Finally, our analysis, which also allows the identification of monetary policy and aggregate demand shocks, suggests that demand and supply factors are almost equally responsible for the movements in the inflation rate during the pandemic.
…
On the demand side, some scholars attribute a considerable portion of the rise in inflation to monetary and fiscal policy mistakes. Reis (2022) and Summers (2022) emphasize the cost of a delayed monetary policy response to increases in inflation. Jord`a et al. (2022), Jord`a and Nechio (2022) and Summers (2021) underline the role of fiscal expansions in the sharp increase in inflation. Our results highlight that monetary policy significantly contributes to inflation, especially in 2021 during which authorities use the “transitory” language in their guidance. We also find the aggregate demand increasingly contributes to the rise in inflation following the month when the second fiscal stimulus is announced under the Biden administration in 2021. Overall, in our framework, demand factors including monetary policy shocks are found to contribute to half of the rise in US inflation

The BoC has released a Staff Discussion Paper by Sarah Miller and Patrick Sabourin titled What consistent responses on future inflation by consumers can reveal:

Inflation expectations play a vital role in determining inflation. Central bankers need to understand their intricacies and the information they can reveal. We look at the consistency of consumers’ answers to questions on inflation expectations in the Bank of Canada’s Canadian Survey of Consumer Expectations. We analyze factors that may explain consistencies among individuals and overall. We also compare the inflation forecasts of consumers with consistent responses with those of professional forecasters and consumers with varying responses.
…
Consumers aged 30 years and over with income above $100,000 and a post-secondary degree are more likely to provide consistent answers. The most important characteristic for consistency is age. For instance, respondents aged 55 years and over are almost twice as likely to provide consistent answers than respondents under 30 years old. We also find that women are less likely to be consistent than men—their point predictions exceed their subjective distributions more often. Indigenous people are less likely than non-Indigenous people to respond consistently. Similar to the gap between consumers’ perceptions of inflation and actual inflation, the likelihood that consumers provide consistent responses may also vary by their shopping patterns.5 For instance, they may be influenced by the prices of goods they purchase frequently, such as food and gasoline. Several factors may influence why certain demographic groups are more likely than others to provide consistent responses. These factors include:

•  a propensity for overpredicting because it is less costly than underpredicting (or vice versa). This is known as asymmetric loss and may cause a consumer to report a point forecast that is outside the interval from their subjective probabilities (for more details, see Patton and Timmermann 2007).
•  a tendency to round probabilities, although Engelberg, Manski and Williams (2009) tested and rejected this hypothesis
•  the possibility that youth, with less experience of high inflation, update their inflation expectations more strongly in response to surprise inflation than older people do (proposed by Malmendier and Nagel 2016)
•  other unobserved personal characteristics such as patience and maturity

…
To determine whether the inflation expectations of consumers with consistent responses are more informative of actual CPI inflation than those of other consumers, we look at perceived inflation (i.e., inflation over the past 12 months) separately for these two groups. Chart 7 reveals striking differences. Perceived inflation averaged across consumers with consistent responses is close to actual CPI, but well above actual CPI for consumers with inconsistent responses. Instead, the pattern of perceived inflation averaged across the inconsistent group is closer to CPI inflation for gasoline.
…
We know that consumers with consistent views are better than other consumers at forecasting inflation. But how does their forecasting compare with that of professional forecasters? Before the pandemic, the one-year-ahead inflation expectations of consumers with consistent views were systematically above actual inflation four quarters later (Chart 8). Forecasting errors for these consumers were larger than those of professional forecasters in Canada surveyed by Consensus Economics.10 Since the pandemic started, forecasting performance has deteriorated for both groups. However, the one-year-ahead inflation expectations of consistent consumers have been closer to actual inflation than those of professional forecasters. This suggests that, when assessing risks around inflation, we should seriously consider the survey results from these consumers. Their expectations represent an upside risk to the inflation outlook presented in the January 2023 Monetary Policy Report.
…
However, like Braitsch and Mitchell (2022), we think targeted and additional forms of communication may be needed to reach consumers with inconsistent responses to encourage them to moderate their inflation expectations. The Bank’s The Economy, Plain and Simple series and the Bank of Canada Museum’s education programs are steps in that direction.

And the BoC has released (boy, they were busy today!) a Staff Discussion Paper by Lin Chen and Stephanie Houle titled Turning Words into Numbers: Measuring News Media Coverage of Shortages:

We generate high-frequency and up-to-date indicators to monitor news media coverage of supply (raw, intermediate and final goods) and labour shortages in Canada. We use natural language processing to construct two news-based indicators and time-varying topic narratives to track Canadian media coverage of these shortages from 2000 to 2022. This makes our indicators an insightful alternative monitoring tool for policy. Notably, our indicators track well with monthly price indexes and measures from the Bank of Canada’s Business Outlook Survey, and they are highly correlated with commonly tracked indicators of supply constraint. Moreover, the news-based indicators reflect the attention of the public on pressing issues.
…
We use the Cision news media database to build a corpus (a collection of written texts) of Canadian English economic and management news articles published between January 1, 2000, and July 4, 2022. We extract sentences containing the keyword “shortage” and collect them into daily documents. Then, we use unsupervised natural language processing (NLP) to decompose the news corpus into different shortage topics. These topics reflect media attention on, among others, the supply and labour shortages that affected Canada’s economy from the early 2000s to the 2020–22 period of the COVID-19 pandemic.

SEC Commissioner Caroline A. Crenshaw gave a speech titled Fixed Income and Options: The Other Market Structures:

The corporate bond and municipal securities markets are relied upon by both retail and institutional investors, including Americans who are approaching retirement or are already there. In the corporate bond market, trades under $100,000 account for between 60% and 70% of reported customer transactions.[17] In the municipal securities market, transactions of less than $25,000 account for more than half of the trades, and those less than $100,000 account for 87% of trades, reflecting that individual investors hold the majority of outstanding municipal bonds.[18]

Investors in these markets are incurring trading costs that far outstrip the costs of transacting in the equity markets. While estimates of trading costs can be a challenge in part due to the relative lack of transparency, academics have estimated corporate bond trading costs at around 84 basis points[19] and municipal bond trading costs as high as 90 basis points for retail-size trades.[20] Surprisingly, smaller bond transactions, which are more likely to originate from retail investors, are more expensive to complete than larger transactions – the opposite of the pattern typically observed in equity markets.[21]

One way to reduce transaction costs and improve investor outcomes would be to improve price transparency. Post-trade price transparency, via TRACE and EMMA, has been a feature of the U.S. fixed income markets, to varying degrees, since the 1990s, and there have been significant improvements over the last several years. However, fixed income markets still largely lack the pre-trade price transparency that has been a feature of the equity markets for decades. While there are some quotation data available from dealers and electronic venues, smaller dealers are less likely to have access to these data or the ability to consolidate them effectively, and they are generally not visible to the retail customer and therefore cannot be used to help the customer negotiate a better price with its dealer.[22] And post-trade information for infrequently traded bonds can be stale or even unavailable.[23] Consistent with this, research on municipal bond markets from the SEC’s Division of Economic and Risk Analysis showed that the majority of customer trades execute at worse prices than best available dealer quotes.[24]

The Corporate Bond Market Distress Index was updated:

• Corporate bond market functioning appears healthy, with the overall market-level CMDI remaining stable around its historical 25th percentile.
• Market functioning in the investment-grade segment has continued to improve in March. The high-yield segment is slightly up, but still at low historical levels.

BIS has released a working paper by Ernest Dautović, Leonardo Gambacorta and Alessio Reghezza titled Supervisory policy stimulus: evidence from the euro area dividend recommendation:

Focus
In March 2020, the European Central Bank made the recommendation that, at least until October 2020, no “significant institution” should pay out dividends. We investigate the recommendation’s impact on the credit supply to non-financial corporations amid the Covid-19 economic shock. Bank managements effectively faced a choice of how to allocate their capital when deciding whether to follow the ECB recommendation, with differing implications for the credit supply. On the one hand, given constant demand and price effects, they might have opted to use the surplus capital to increase lending supply, thus responding countercyclically to support the economy. On the other hand, they might have decided to increase their resilience to future shocks by saving capital, and/or strengthening their loss-absorption capacity by making additional provisions. The paper asks whether the ECB’s dividend recommendation led to an increase or a decrease in the credit supply to non-financial corporations, and whether this effect varied for different types of firm and sector.

Contribution
The study compares the credit supply of banks affected by the ECB recommendation with a group of unaffected banks, and controls for other pandemic-related support measures. To address identification issues, we rely on credit registry data and a direct measure that captures differences in compliance with the dividend recommendation across banks in the euro area. The analysis disentangles the confounding effects stemming from the wide range of monetary and fiscal policies that supported credit during the Covid-19 downturn and investigates their interaction with the dividend recommendation.

Findings
We find that dividend restrictions have been an effective policy in supporting financially constrained firms, adding capital space to banks, and restricting some forms of procyclical behaviour. In particular, the study finds that the dividend recommendation added 4.4 percentage points to the growth rate of euro area credit supply to non-financial corporations. The effects on lending are larger for small and medium enterprises and for firms operating in sectors that were exposed to the effects of Covid-19. We also find evidence that the dividend recommendation has sustained bank lending even in the absence of government guarantees. At the same time, we do not find evidence of a significant increase in lending to riskier borrowers and “zombie” firms.

Abstract
At the onset of the Covid-19 outbreak, central banks and supervisors introduced dividend restrictions as a new policy instrument aimed at supporting lending to the real economy and strengthening banks’ capacity to absorb losses. In this paper we estimate the impact of the ECB’s dividend recommendation on bank lending and risk-taking. To address identification issues, we rely on credit registry data and a direct measure that captures variation in compliance with the recommendation across banks in the euro area. The analysis disentangles the confounding effects stemming from the wide range of monetary and fiscal policies that supported credit during the Covid-19 downturn and investigates their interaction with the dividend recommendation. We find that dividend restrictions have been an effective policy in supporting financially constrained firms, adding capital space to banks, and limiting some forms of procyclical behaviour. The effects on lending are larger for small and medium enterprises and for firms operating in Covid-19 vulnerable sectors. At the same time, we do not find evidence of a significant increase in lending to riskier borrowers and “zombie” firms.

I must admit, this kind of research makes me nervous. Look at the abstract! “… central banks and supervisors introduced dividend restrictions as a new policy instrument aimed at supporting lending to the real economy and strengthening banks’ capacity to absorb losses.” OK, the second objective is well within the aegis of the supervisory class, but the first? That’s just mission creep. If the authorities want to boost lending, they have lots of less invasive tools in the box already.

And in Canada, thirty year mortgages will become more common:

Proposals from the Financial Consumer Agency of Canada are aimed at ensuring fairness and consistency in terms of relief offered for struggling borrowers. The plan was highlighted in the federal budget this week in a clear sign that Ottawa endorses the idea of allowing mortgages to grow to keep payments down.
…
The FCAC said it developed the guidelines for mortgage borrowers at risk of missing monthly payments because of what it called “exceptional circumstances.” The FCAC did not provide a measurable definition of “exceptional circumstances” except to say it could be the combined effects of high household debt, rapid interest-rate hikes and higher cost of living.

Under the proposal, banks could lengthen their troubled borrowers’ amortization periods either temporarily or permanently. The guideline would apply to all types of mortgages, including those with a fixed interest rate.

If it is temporary, the watchdog said banks should take into account the borrower’s ability to restore the amortization to the original period within what it said was a “reasonable time frame.” If it is a permanent solution, the bank must ensure the amortization period is “reasonable.” FCAC would not comment on what it considered reasonable.

PerpetualDiscounts now yield 6.25%, equivalent to 8.12% interest at the standard equivalency factor of 1.3x. Long corporates yielded 4.95% on 2023-3-24 and since then the closing price has changed from 15.18 to 14.98, a decrease of 132bp in price, with a Duration of 12.41 (BMO doesn’t specify whether this is Macaulay or Modified Duration; I will assume Modified) which implies an increase in yield of about 11bp since 3/24 to 5.06%, so the pre-tax interest-equivalent spread (in this context, the “Seniority Spread”) has narrowed substantially to about 305bp from the 330bp reported March 22.