Graph of the week

Graph of the week

12-07-2018 | インサイト

A costly alarm clock

  • Peter van der Welle
    van der Welle

This month marks the second-longest economic expansion phase in US post-war history. With the ageing of the cycle, the supply of recession indicators calling for an end to this phase in the business cycle has increased accordingly. These indicators are clearly in demand as some investors observe that the length of this expansion phase, which started in 2009, now far exceeds the average, suggesting the expansion must be bound to die of old age soon.

The graph I stumbled upon introduces yet another, lesser-known recession indicator. It shows the compensation investors demand for investing in investment grade corporate bonds in excess of the yield on risk-free US government bonds, which is now around 2%. These BAA-rated bonds are subject to moderate credit risk. They are a pretty good gauge of the average credit risk in the market, as investors in this segment do not have a lot of protection, though their investments are not poorly secured, either. Normally, the issuers are considered highly likely to repay the debt, but in contrast to AAA-rated bonds, certain protective features are lacking.

The message that goes with this graph is that the recent breach of the 2% credit spread level sends a dire signal that should not be ignored, as it could point to an upcoming recession. But is this really a signal or is it just a bit of noise that we’re actually better off ignoring? The graph’s creators claim that a credit spread exceeding 2% has a good track record of predicting a recession.

But why 2%? If this is the level that spells disaster, then why is it that the indicator has continually been flashing red since the onset of this expansion phase? A very expensive false alarm indeed with a US stock market that has gone up 213% during the same period. If it is the breach of 2% itself that is cause for alarm, then why was there no recession after spreads exceeded that level in 1991, 1992 and 1998? Only in 1989, 2000 and 2007 was a breach of 2% followed by a recession within the subsequent six to 12 months. This creates a powerful framing effect. Unfortunately, a hit ratio of 50% over the past 32 years is not a good track record and hardly convincing as a guide for navigating markets. Put simply: it’s no better than flipping a coin.

Of course, it is true that higher BAA credit spreads typically correlate with a higher estimated probability of corporate default, which could forebode economic contraction. As such, the BAA spread is a useful factor in a model that tracks the business cycle. However, to say it can predict shifts in a business cycle is quite a leap. Without the proper theoretical underpinning or economic intuition that would explain what’s so special about the 2% spread level in relation to a heightened risk of recession, this graph is of little value as a recession indicator and could prove a costly alarm clock.


当資料は情報提供を目的として、Robeco Institutional Asset Management B.V.が作成した英文資料、もしくはその英文資料をロベコ・ジャパン株式会社が翻訳したものです。資料中の個別の金融商品の売買の勧誘や推奨等を目的とするものではありません。記載された情報は十分信頼できるものであると考えておりますが、その正確性、完全性を保証するものではありません。意見や見通しはあくまで作成日における弊社の判断に基づくものであり、今後予告なしに変更されることがあります。運用状況、市場動向、意見等は、過去の一時点あるいは過去の一定期間についてのものであり、過去の実績は将来の運用成果を保証または示唆するものではありません。また、記載された投資方針・戦略等は全ての投資家の皆様に適合するとは限りません。当資料は法律、税務、会計面での助言の提供を意図するものではありません。




商号等: ロベコ・ジャパン株式会社  金融商品取引業者 関東財務局長(金商)第2780号

加入協会: 一般社団法人 日本投資顧問業協会