始まりは終わり、終わりは始まり：景気後退の見え方

この記事は投資人生を通じて有効です。

The Beginning is the End is the Beginning: A Look at Recessions

Market Commentary

Stocks

By Liz Ann Sonders

February 25, 2019

Key Points

• Recessions are not back-to-back negative GDP quarters; they’re instead defined using four key coincident economic indicators.
  
  景気後退とはマイナスGDP成長四半期が連続することではない；そうではなく４つの重要経済指標で定義される。
• Bear markets often overlap with recessions (and typically lead them), but not always.
  
  ベア相場は景気後退と重なることが多い（典型的には先行する）、しかし必ずそうというわけでもない。
• It’s important to distinguish between leading and lagging economic indicators and to focus at least as much on trend as level.
  
  大切なことは先行指標と遅行指標を見極めること、そしてすくなくともそのトレンドに注意せよ。

Recession chatter is abundant lately. It’s increasingly the focus of Q&A sessions at investor events at which I’ve been speaking. I also received a series of questions last week about recessions from a Schwab colleague who has many younger Schwabbies on his team, most of whom have not lived as working adults through a recession. In putting together answers to his questions in one of our internal sites, I decided it was a topic to which I should devote these pages.
Perhaps heightened recession concerns are to be expected given the duration of the current cycle—which will become the longest post-WWII expansion if a recession doesn’t begin by July of this year. Or perhaps it’s because of the recent deterioration in economic data across a fairly wide spectrum of indicators. I’ve been touching on the topic quite a bit in my writings as well as on Twitter, but it’s time for a more evergreen look at recessions.

景気後退の話題が喧しくなって来た。私の講演会でもQ&Aでこの質問が増えている。先週にはSchwab社の同僚からもこの手の一連の質問を受けた、彼は多くの若手社員をチームとして抱えている、彼らの多くは働きだしてから景気後退を経験したことがない。その時の返答をまとめて社内ネットに掲載したが、皆さんにも知ってもらおうとこの記事を書いた。たぶん、景気後退懸念の高まりは今回の景気拡大期間の長さによるーー今年７月になると戦後最長の景気拡大期となる。もしくは最近の経済データが広範囲に悪化しているからかもしれない。私はこの件に関して何件かの書物もしたしTwitterでも書いてきた、しかし景気後退に対してもっと長期的視点で議論する良い時期だと感じている。

Inevitability　必然

The bottom line is the U.S. economy will move into a recession at some point. It’s inevitable. They always occur at the end of a cycle and set the stage for the subsequent cycle. Recessions haven’t been outlawed, nor can (or should) they be prevented at all costs by the Federal Reserve or other policy-makers. What we don’t know is the length of runway between now and the next recession. I’ve been positing that at this stage, an earnings recession seems more likely in the near-term (i.e., starting sometime in this year’s first half) than an economic recession. But it’s never too early to refresh our memories as to what to look for to gauge the risk and timing of recessions.


確実なことはどこかの時点で米国にも景気後退がやって来るということだ。これは回避不能の必然だ。それは景気サイクルの最後に到来し、その後次の景気サイクルが始まる。景気後退は決して否定されるものではなく、FEDや政権の力を持ってしても回避できるものではない。我々がわからないのは次の景気後退までの期間だ。現時点において、収益景気後退がとても近いと私は思っている（今年前半の何処かで始まると思っている）、経済的景気後退というわけではない。しかし景気後退のリスクとタイミングについて思い起こしてみるのに早すぎることは決してない。

What a recession is and isn’t

何が景気後退で何がそうではないか

First, let’s get the definition straight. I’m always surprised when I hear or read the perceived definition of a recession being two consecutive quarters of negative gross domestic product (GDP). That is not, nor has ever been, the definition of a recession. The official arbiter of recessions is the National Bureau of Economic Research (NBER) and they define a recession as “a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales.” Not coincidentally (pun intended), those latter four economic readings make up The Conference Board’s Index of Coincident Indicators, which get released monthly alongside the Leading Economic Index (LEI), which I’ll get to later in this report.

まずはその定義から始めよう。私はいつも驚くのだが、２四半期連続でGDP成長がマイナスのことだと定義する人がいる。これは決して景気後退の定義ではない。公式に景気後退を判定するのはNBERであり、彼らはこう定義している「経済活動全般に深刻な後退が見られ、これが数ヶ月続くこと、通常はGDP、収入、雇用、工業生産、そして卸売り小売に現れる」。必ずしも同時でなくてよいが、これら４経済指標は The Conference Board's Index of Coincident Indicators　と呼ばれる、この値は毎月 Leading Economic Index(LEI)として開示される、この記事の後半でこの議論をしよう。


For “proof” of the difference between the perceived definition and the actual definition, note the following:

世間で言われる定義と実際の定義の違いの「実例」はこういう具合だ：

• The 2001 recession did not have back-to-back negative GDP quarters.
  
  ２００１年の景気後退では連続四半期でのマイナスGDP成長はなかった。
• The 1960-1961 recession did not have back-to-back negative GDP quarters.
  
  １９６０−１９６１景気後退でも連続四半期でのマイナスGDP成長はなかった。
• There were back-to-back negative GDP quarters in 1947, yet there was no recession.
  
  １９４７年には連続してGDPがマイナス成長だった、しかし景気後退ではなかった。

Below is the full post-Great Depression roster of official recessions.

下に示すのは大恐慌以来のすべての公式景気後退だ。

Source: Charles Schwab, National Bureau of Economic Research (NBER). *Back-to-back “double dip” recessions.

As you can see, the average span of recessions during this time period was 10.8 months, with a range from six months to 18 months. Keep in mind, that the NBER is generally late in declaring the start and end dates of recessions—they do not do any recession forecasting, but instead wait until the trends are clear with regard to the parameters they use for declaration and dating purposes.

見ての通り、この時代の平均景気後退期間は１０．８か月だった、範囲としては６か月から１８か月。覚えておくが良いが、NBERが景気後退の開始や終了を宣言するのはあとになって振り返ってだーー彼らは決して景気後退を予想はしない、彼らの注目する指標のトレンドが明確になってからのことだ。

Stock market relative to recessions

景気後退と株式相場

What should we expect from the stock market during a recession? Should we assume 20% or more market losses given the tendency for stocks to enter into a bear market in anticipation of recessions?


景気後退期に株式市場になにを期待すればよいだろうか？景気後退懸念でベア相場いりし２０％かそれ以上の下落を予想すべきだろうか？

The table below shows every S&P 500 bear market (using the traditional -20% definition) in the post-Great Depression era, but also what I’ll call “near bear markets” (down at least 19% but less than 20%).

下の表はS&P500のベア相場（２０％以上の下落）を示している、大恐慌以来のものだ、しかし私は「疑似ベア相場」（２０％に満たないが１９％を超える下落）も加えた。

As you can see in the table below, there is typically an overlap between the two, with bear markets generally starting in advance of recessions (red entries are bear/near bear markets which overlapped with recessions). That said, there are bear/near bear markets that have not accompanied recessions (like 1987); and there are recessions that have only had near bear markets (like 1991). There are a surprisingly large number of near bear markets.

下の表を見ればわかるが、２つが重なることがある、景気後退入りする前にベア相場となる場合（景気後退を伴う場合を赤文字）だ。景気後退を伴わない場合もある（たとえば１９８７年）；景気後退だが疑似ベア相場でしかない場合（たとえば１９９１年）。驚くことに擬似ベア相場というのがたくさんある。

Source: Charles Schwab, Bloomberg, National Bureau of Economic Research. 

Bear market defined as 20% or greater drop in S&P 500.  Near bear market defined declines of more than 19% but less than 20%. *3/24/2000–10/9/2002 is generally considered one long bear market (-49.1%), but there were two 20% rallies within that span. **10/9/2007 – 3/9/2009 is generally considered one long bear market (-56.8%), but there was one 20% rally within that span.

S&P500が２０％以上の下落をするとベア相場と呼ばれる。疑似ベア相場とは１９％以上だが２０％に満たない場合だ。＊３／２４／２０００−１０／９／２００２は長いベア相場だがその何は二回の２０％のラリーがあった。＊＊１０／９／２００７ー３／９／２００９は長いベア相場だが、その期間に一度２０％のラリーがあった。

Bear markets that overlapped with recessions were generally more severe than those occurring outside recessions. The average bear/near bear market without a recession was -24.6%, while the average bear/near bear market with a recession was -32.2%.

景気後退を伴うベア相場はより厳しいものだ。景気後退を伴わないベア疑似ベア相場の平均下落は−２４．６％だが、景気後退を伴ったときは−３２．２％にもなる。

My old adage　私の好みの格言

This next section will highlight several popular economic indicators—and the importance of understanding the difference between lagging and leading economic indicators. It will also help to illustrate my well-worn, and oft-cited adage that “when it comes to the relationship between economic data and the stock market, better or worse tends to matter more than good or bad.” In other words, we often think of what’s happening in the economy in “good or bad” or “strong or weak” terms; when we should be thinking in “better or worse” terms.

次にいくつかの経済指標について議論しようーーそこで先行指標と遅行指標の違いの重要性を指摘する。これを見るとよく言われる陳腐な格言を示している「経済データや株式市場に関しては、良いか悪いかの絶対値ではなく改善、悪化トレンドが重要だ」。言い換えると、我々は度々経済現象に対して「良いか悪いか」とか「強いか弱いか」を口にする；本来は「改善しているか悪化しているか」というべきだ。

Unemployment rate “too low” to suggest increasing recession risk?

失業率が「とても低い」というのは景気後退リスクを示唆しているだろうか？

When discussing recessions with investors, and listening to many pundits in the media, I often hear today’s low unemployment rate (UR) cited as a reason not to fret a recession any time in the near future. But here’s the rub—the unemployment rate is one of the most lagging of all economic indicators. In fact, as you will see in the chart below, the unemployment rate has always been low at the outset of recessions. It’s most significant period of deterioration has historically been during recessions; not in the lead-up to recessions. Put another way, a rising unemployment rate doesn’t cause recessions; recessions cause the unemployment rate to rise.

投資家が景気後退を議論するとき、またメディアのアナリストの話を耳にするとき、私は現在の低失業率をもって近い将来の景気後退を気にすることはないと言われる。しかし底に誤解があるーー全ての経済指標の中で失業率は最も遅れた遅行指標なのだ。実際下のチャートを見れば解るだろうが、景気後退の始まりでは常に失業率は低い。これが悪化するのは常に景気後退期なのだ；決して景気後退に先行することはない。言い換えるなら、失業率上昇が景気後退の原因ではない；景気後退が失業率を増やすのだ。

Unemployment Rate Around Recessions

Source: Charles Schwab, Department of Labor, FactSet, Ned Davis Research, Inc. (Further distribution prohibited without prior permission. Copyright 2019(c) Ned Davis Research, Inc. All rights reserved.), as of January 31, 2019.

In the post-WWII era, the average uptick in the UR from its trough to the month the recession began has been 0.4%, with a range of 0.0% to 0.7%. In other words, there were times when the UR hadn’t moved up at all, yet a recession was ultimately declared as having started.

第二次世界大戦後、失業率の底値から景気後退開始までの増加は平均０．４％だった、レンジとしては０．０％から０．７％だ。言い換えると、失業率が全く増えなくとも景気後退を宣言されることもあった。

It’s also instructive to see the data in the table above, which highlights what happens when you compare a highly lagging indicator (the UR) to a highly leading indicator (the stock market). The worst returns for the stock market historically came when the UR was in its lowest zone (think last year); while the best returns have come when the UR was in its highest zone (think the start of the bull market in March 2009). This is because the stock market generally anticipates the coming increase in the UR once it’s sniffed out a recession; and also anticipates the coming decrease in the UR once it’s sniffed out a recovery.


上の表の数字が教えてくれるが、遅行指数である失業率や先行指標の株式市場で何が起きるかだ。歴史的に見て失業率が最低域（昨年を思い起こすが良い）では株式のリターンは最悪だ；一方で株式の最良リターンが得られるのは失業率が最高のとき（２００９年３月のブル相場の始まりを考えると良い）。

（訳注：２００９年３月６日（金）の新聞記事の一面トップは失業の記事です。第一金曜で雇用統計発表日です。ここが底値でした。後から振り返るととても解りやすい、でもそのときにはだれもが悲観的だったのです。
https://money.cnn.com/2009/03/06/news/economy/jobs_february/　）

Unemployment claims troughs lead recessions

失業率が底を打つと景気後退が始まる

Although less “popular” as an indicator, the most leading of the various employment indicators is initial unemployment claims. In fact, they’re one of the 10 subcomponents of the LEI, which I’ll get to shortly. The fact that claims recently broke out to the upside suggests we need to be on guard for the signal they’re giving about the length of time between now and the next recession. If they continue to tick higher, the risk of a recession starting sooner rather than later will move up.

あまり「一般的」な経済指標ではないが、各種雇用指標のなかで最も先行指標となるのは初回失業保険申請指数だ。実際この指数はLEIの１０要素の一つになっている、あとで簡単に議論しよう。この申請が最近上昇している、ということは次の景気後退までの猶予を気にしなければならない。もしこの指数が引き続き上昇を続けるなら、景気後退開始リスクは早まっている。

Unemployment Claims Around Recessions

景気後退前後の失業保険新規受給申請件数

Source:  Charles Schwab, Department of Labor, FactSet, as of February 15, 2019.

Consumer confidence peaks lead recessions

消費者信頼指数ピークは景気後退につながる

In addition to hearing the unemployment rate cited as a reason not to fret a recession, I often hear the same about consumer confidence. As you can see in the chart below of The Conference Board’s measure of consumer confidence, it remains high in level terms, but is clearly off the peak. Consumer confidence, as a leading indicator, has typically peaked out not too far in advance of recessions’ starts.

失業率が低いから景気後退は心配するな、という話に加えて、私が度々耳にするのは消費者信頼指数にかんする同様の議論だ。下のチャートに消費者信頼指数を示すが、そのレベル自体は高いが、しかし明らかにピークを超えている。消費者信頼指数も先行指数であり、景気後退が始まる前にピークを打つことが多い。

Consumer Confidence Around Recessions

景気後退前後の消費者信頼指数

Source: Charles Schwab, FactSet, The Conference Board, as of January 31, 2019.

Another consumer confidence-related indicator for recession risk is the spread between the “present situation” component of the consumer confidence survey and the “future expectations” component. As you can see in the chart below, extreme troughs in this spread have been consistent indicators of coming recessions. (Of course, we don’t yet know whether we’re at or near the trough for this cycle, but the currently-extreme spread bears watching.)

景気後退リスクに関するもう一つの消費者信頼指数関連指数は「現状」と「将来期待」の乖離だ。下のチャートに示すが、このスプレッドが極端な谷を形成したとき景気後退接近警報となる。（当然のことながら、今どこに位置するか今サイクルの谷にどこまで近づいているかはわからない、しかし現在極端に乖離しているので要注意だ。）


Consumer Confidence Spread Around Recessions

景気後退前後の消費者信頼指数スプレッド

Source: Charles Schwab, FactSet, The Conference Board, as of January 31, 2019.

LEI peak?　LEIはピーク？

I’ve touched on a couple of the more leading indicators, but let’s look more broadly at the full set. I focus on the Leading Economic Index (LEI), put out by The Conference Board. As you can see in the full-history chart below—in their presently-constituted form (they’ve been “back-fitted” to account for changes to the indicators that are most highly-correlated to the business cycle)—the span between LEI peaks and recession starts has been 13 months, with a range of eight to 21 months. For what it’s worth, the span between LEI troughs and recession ends has been only two months, with a range of zero-to-five months; so there is generally more signal lead time heading into recessions than heading out of them.

これまですでに先行指標二件について議論した、しかし全体を網羅するもっと広範囲な議論をしよう。私はLeading Economic Index(LEI)に注目している、The Conference Boardが算出するものだ。下の長期チャートを見ればわかることだがーー現在のモデルにおいては（この指数は過去の景気サイクルとの相関が高くなるように見直されて。この指数の良いところはLEIの谷から景気後退終了までの期間がわずか２か月ということろだ、レンジは０−５か月だ；そのために景気後退脱出予想よりも景気後退入り予想のほうがリードタイムが長い。


Leading Indicators Around Recessions

景気後退前後の Leading Indicators 

Source: Charles Schwab, FactSet, The Conference Board, as of January 31, 2019.

In its presently-constituted form, the LEI never failed to give a heads up that a recession was coming. For what it’s worth, the LEI peaked last September and declined in two of the subsequent four months. It’s perhaps too soon to judge whether last September’s peak was the peak for the cycle, especially given the temporary effects of the government shutdown, but we’ll see.

現在の計算式において、LEIはこれまで景気後退到来を見逃したことはない。だからこそこの指数は有用だ、このLEIが昨年９月にピークをつけ、その後４ヶ月のうち二ヶ月で下落している。昨年９月のピークが今サイクルの本当のピークかどうか判断するのはまだ早すぎるが、特にその後一時的政府封鎖があったことを考慮すると、注視せざるを得ない。

Looking under the LEI’s hood, you can also see where the deterioration has been concentrated. I also included a Coincident Economic Index (CEI) section given the aforementioned connection between those indicators and the NBER definition of recessions. It’s true that there is not yet any “red” flashing for these indicators in level terms; but there certainly is some flashing occurring if you look at the indicators’ trends.

LEIを通して見通すと、経済のどの分野に劣化が起きているのかがよく分かる。私はまたCoincident Economic Index(CEI)も取り上げよう、この指数とNBERが定義する景気後退との関係だ。たしかにこの指数をみてもそのレベル（絶対値）に決して「赤信号」はない；しかしこの指数のトレンドを見るとたしかに何らかの警告が点灯していることはたしかだ。

Source: Charles Schwab, FactSet, The Conference Board, as of January 31, 2019.

Election year　選挙年

One of the aforementioned questions I received from my colleague was about whether a recession is possible during an election year (yes, I know the election isn’t until next year). Historically, recessions have sometimes occurred during election years:


最初に書いたが同僚から受けた質問の一つが、選挙年に景気後退があるかどうかというものだった（そう、来年が選挙年だ）。歴史を振り返ると、景気後退が選挙年に起きるのも度々だ：

• In 1932, we were in the midst of the Great Depression (1929-1933). 
• In 1948, a recession began in the same exact month as that year’s election. 
• In 1960, a recession began early that year. 
• In 1980, we were already in the “double-dip” recession(s) ending in 1982. 
• In 2008, we were already in the midst of the worst recession since the Great Depression.

Fed models say what?

FEDの景気モデルではどうなっているだろう？

The Federal Reserve has not distinguished itself historically with forecasting recessions. But that doesn’t mean they haven’t created forecasting models for both recessions and GDP growth. Among myriad recession probability models out there, the one from the Federal Reserve Bank of New York is fairly popular. As you can see in the chart below, it’s showing a 24% chance of a recession, which doesn’t sound high. But as shown, with the exception of the late-1960s and mid-1990s, once the model reached that level it continued to rise and recessions were soon on the way.

歴史的にみて、FEDは景気後退を予見する名誉は得てない。しかしだからといって彼らも景気後退やGDP成長モデルを作ってないわけではない。景気後退予想モデルは無数にあるが、NewYork FEDのモデルはよく知られている。下のチャートに示すが、現在の景気後退可能性は２４％だ、まだそう高くない。しかし下のチャートを見れば明らかだが、１９６０年代後半と１９９０年代なかばの例外を除くと、このモデルであるレベルを超えると上昇を続けすぐに景気後退が差し迫る。

Recession Probability Model

Source: Charles Schwab, Federal Reserve Bank of New York, as of January 31, 2019. Model uses difference between 10-year and 3-month Treasury rates to calculate probability of a recession 12 months ahead.

Both the Atlanta and New York Federal Reserve Banks also publish GDP forecasting models, which you can see below. Atlanta’s model, called GDPNow, doesn’t yet have a new forecast for this year’s first quarter, but their forecast for 2018’s fourth quarter dropped precipitously from 3% late-last year, but recently rebounding to 1.9% today. The NY Fed’s model, called Nowcast, does have a forecast for this year’s first quarter and it’s dropped from 2.6% late-last year to only 1.2% today (although their fourth quarter 2018 forecast is 2.3%, so higher than GDPNow’s).

Atlanta FEDとNewYork FEDがともにGDP予想モデルを開示している、下に示すとおりだ。AtlantaモデルはGDPNowと呼ばれる、まだ今年Q1の新規予想を開示していない、しかし彼らの２０１９Q4予想は昨年遅くの予想３％から急落し、その後持ち直して今日の時点で１．９％だ。NY FedモデルはNowcastと呼ばれ、今年Q1の予想が昨年遅くの２．６％から急落して今や１．２％しかない（ただし２０１８Q4予想は２．３％でGDPNow
より大きい）。

GDPNow for 4Q2018

Nowcast for 1Q2019

Source: Charles Schwab, Bloomberg, Federal Reserve Bank of New York, as of February 22, 2019.

The mother of all recession indicators

全ての景気後退指標の根拠

We saved the best for last and will conclude with a look at what has arguably been the best recession forecasting indicator historically—the yield spread. There are myriad spreads across the Treasury duration spectrum, but the one historically most useful for forecasting recessions is the spread between the 10-year and three-month Treasury yields. As you can see in the chart below, inverted yield curves (when long-term rates fall below short-term rates) has generally been followed shortly thereafter by recessions.


いちばん大切なことを議論の最後まで残してきた、歴史的に見て最良の景気後退指標に関する結論を言おうーーイールドスプレッドだ。米国債の償還期間に応じて多くのスプレッドがある、しかし景気後退予想で最もよく使われるのは１０年債と３か月もの財務省証券の金利差だ。下のチャートに示すように、イールドカーブが反転するとすぐ後に景気後退となる。

Yield Curve Around Recessions


Source: Charles Schwab, FactSet, as of February 22, 2019.

More recently, in FEDS Working Paper No. 2018-055, Fed Governor Eric Engstrom and Fed Research & Statistics Economist Steven A. Sharpe argued that the spread of short-term Treasury rates—the difference between the six-quarters-ahead forward rate and the three-month yield—might be preferable as a predictor because it focuses on expectations of the near-term path of monetary policy. For what it’s worth, that spread did invert briefly at the beginning of January this year.

つい最近、FED論文 No2018-055で、FED総裁 Eric EngstromとFEDエコノミスト Steven A.Sharpeがこう主張した、短期国債金利スプレッドーー6四半期forward rate と３か月もの国債金利差ーーこれが景気後退予想に有用だと、というのもこれらを見ることで短期的な金融政策動向を推測できると。実際これがどれほど有用かと言うと、今年１月初めに短期的にこのスプレッドが反転した。

Concluding with hope　希望を込めた結末

Having little interest in being a Debbie Downer, I’ll conclude with the reasons why the length of runway between now and the next recession could remain fairly long. Both household and business balance sheets remain relatively healthy. The Fed has taken a dovish turn. Some version of a trade deal may be in the works. And although still up from last year’s low, unemployment claims have recently ticked lower again. But it’s never too early to spend time assessing the risks associated with recessions. Perhaps last year’s near bear market was an indication of a recession starting as soon as this year, or perhaps it was a head fake. I believe an earnings recession is a possibility this year, even if we can avoid an economic recession starting this year. But given myriad late-cycle conditions, we continue to urge investors to be prepared for the end of this cycle.


議論が盛り上がったところで座をしらけさすようなことは好まないが、どうして次の景気後退まで結構時間がかかるかという理由をもって結語としたい。家計も企業もバランスシートは比較的健全だ。FEDはハト派姿勢に転向した。何らかの形で貿易取引は合意を得るだろう。そして昨年の超低位から多少悪化したが、新規失業率申請は最近また少し減っている。しかし景気後退リスクを考慮する時間をとるのもまだ決して早すぎるわけではない。昨年の準ベア相場下落を見ると多分今年には景気後退意入する兆候だろう、もしかするとフェイントかもしれないが。私自身は、たとえ今年景気後退入りしなくとも、今年収益後退になると確信している。数限りなく景気サイクル終盤状況を示すデータがあり、私どもは引き続き投資家に今景気サイクルの終了に向けて準備するよう促している。

Next Steps

• Follow Liz Ann Sonders on Twitter: @lizannsonders.
• Explore the investment help and guidance Schwab offers.
• Stay connected with the latest investing insights from Schwab.
• Talk to us about the services that are right for you.  Call our investment professionals at 800-435-4000.