Recent Posts

Sunday, February 11, 2018

Introduction To "Inflation Breakeven Analysis"

The great inflation of the 1970s in the developed countries provoked a strong political and economic reaction. Investors searched for ways to protect themselves from inflation. The United Kingdom launched the first modern inflation-linked bonds in 1981. In addition to being of interest to investors looking for protection against inflation, the market also provides a market-based measure of inflation expectations. Since investors have “skin in the game,” the resulting forecasts could be better than a purely survey-based inflation forecast. A more reliable inflation forecast could then be useful for policymakers that aim to control inflation.

This article is the unedited introduction to the upcoming report "Inflation Breakeven Analysis."

I have neared the completion of the first draft of the text. I am putting this article out as an advanced advertisement for what to expect, and as a replacement for my usual weekend article. I have left the text as-is, which means that it refers to "sections" and "report," which does not entirely make sense for an article on a website. As noted, this section is quite rough, and I hope to refine it. Once I finish the first draft, I will put it aside for about a month, so that I can start revising the text with fresh eyes. Unless there is something seriously wrong with the text, it should be released about a month or two after the second revision process is started. 

This report discusses the breakeven inflation rate that is implied by pricing in the fixed income markets. For those with a casual interest in the subject, it is probably good enough to view those inflation breakeven rates as a market-implied forecast for average inflation over a horizon that matches the maturities of the bonds used to calculate the inflation breakeven.

However, if one wants to delve into the analysis, it is necessary to come to grips with the complications in the subject. Is the forecast biased? Are there technical factors in the bond market that affect pricing? The objective of this report is to offer an intermediate-level introduction to these issues. The target audience is either those with an interest in finance and are unfamiliar with inflation-linked bonds, or economists that want to understand better the factors that affect inflation breakeven rates.

The style of writing is more advanced than earlier reports by BondEconomics, reflecting the different target audience. It is meant to provide to act an explanation of what the technical details in the academic literature mean, and is not meant to act like an academic text. Although some equations appear, calculations that are more complex are deferred to the source code (that is available free in an open source Python package).

The scope of this report is deliberately limited. The discussion is focussed on the inflation breakeven rate, what it means, and the factors that influence it. Although some discussion of valuation techniques appears, it certainly does not represent investment advice. Instead, it may help the reader understand the factors that might influence their own investment strategy.

Topics that are discussed include:

  • principles behind inflation-linked calculations;
  • inflation swaps;
  • curve fitting;
  • effects of seasonality;
  • the importance of energy prices;
  • principal puts;
  • inflation risk premium;
  • fundamental valuation techniques.

The examples are largely based on data available from Canada and the United States, along with some time series from the United Kingdom. The author is mainly familiar with the developments in the Canadian and American inflation-linked market, and therefore concentrated the analysis on those markets. The principles discussed are generally applicable elsewhere, although there are sections on the special characteristics of U.K. inflation-linked gilts (Section 2.7) and the euro area (Section 4.8).

Although there is some discussion of the empirical characteristics of inflation, as well as short-term inflation forecasting, the subject of what determines inflation is deliberately set aside. (I insert some opinions about historical events, but they are labelled as such.) Economists are a major component of the target audience for this text, and I do not want to have debates about the economics of inflation distract from the discussion how inflation-linked bonds work. My working assumption is that an economist will tend to have strongly held prior views about inflation, and so there is no point in adding my views in this context. A discussion of the inflation process would have to be a separate book.

Another area that is not dealt with is the precise details of fixed income quote conventions associated with these bonds. If one is developing pricing algorithms (“pricers”) for fixed income, it is necessary to understand exactly these conventions if one is going to get the correct quoted prices and yields for instruments. However, conventions differ across markets, and one has to be very careful about all the details. Instead, the text focuses on highly simplified calculation conventions. For those new to the field, you need to understand how to get the approximate answer before worrying about the day count conventions.

One final area that is skipped over is the question of affine term structure models. These are very popular in academia and in central bank research. I discuss them in general terms in Section 3.5, but do not delve in the mathematics. The author’s opinion is that these models add very little value despite their mathematical complexity. In order to discuss them properly, one would need to discuss multiple models as well as stochastic calculus. Such analysis would be a distraction from the rest of the text.

(c) Brian Romanchuk 2018

No comments:

Post a Comment

Note: Posts may be moderated, and there may be a considerable delay before they appear.

Although I welcome people who disagree with me, please be civil.

Please note that my spam comment filter appears to dislike long "anonymous" posts. I get no warning about this, and only go through my "spambox" infrequently. The best bet it to keep comments short, and if you think the spam filter struck, let me know with a short comment.