Download Tracking the business cycle - JP Morgan Asset Management

MARKET
INSIGHTS
Tracking the business cycle
IN BRIEF
• The length of the U.S. economic expansion is raising concerns
about a recession. But a closer look at past and present
business cycles suggests that the U.S. economy remains in a
mid-cycle stage. The UK and especially the euro area likely
are even earlier in their economic expansions.
• Economic cycles have become longer, probably reflecting
changes in the macroeconomic environment, such as better
monetary policymaking, declines of the manufacturing and
agricultural sectors, the increased size of the public sector
and broader availability of consumer credit.
• While every business cycle is different, certain phenomena
tend to repeat. With many indicators, such as interestsensitive spending, credit growth, and price inflation,
appearing low by long-term standards, the U.S. likely enjoys
scope to grow before recession risk becomes elevated.
Measures of labor market slack, however, are more
uncertain and could play a major role in determining how
the U.S. business cycle unfolds.
• Until a recession becomes imminent, financial markets are
likely to perform reasonably well, with equities moving
higher and credit spreads tightening further. Investors
should tilt toward growth-sensitive assets while, to account
for uncertainty, maintaining robust hedge buckets.
In the middle of 2014, the current U.S.
expansion passed its fifth birthday. This
growth phase has already matched
the average length of the 11 previous
postwar cycles and easily exceeds the
roughly three-year mean for expansions since the mid-1800s.
By long-run standards, then, the expansion looks aged, raising
the question of whether a recession will soon occur. The rapid
decline of the unemployment rate further increases the urgency
of this debate. Given the dramatically different market outcomes observed during expansions and recessions, the likelihood of recession within the next few years represents a crucial
issue for institutional investors.
An examination of business cycle history and a look at the
specific characteristics of the current expansion, however,
provide some reassurance. Expansions have lengthened in the
FOR INSTITUTIONAL/WHOLESALE OR PROFESSIONAL CLIENT USE ONLY | NOT FOR RETAIL DISTRIBUTION
MARKET
INSIGHTS
PORTFOLIO
DISCUSSION:
Title Copy Here
Tracking
the
business cycle
past few decades, probably reflecting changes in the
macroeconomic environment (such as better monetary
policymaking) that are likely to persist. Meanwhile, the current
expansion, characterized thus far by fairly sluggish growth, does
not yet show signs of accumulating the sorts of imbalances that
have been associated with prior recessions. Assuming that the
economy avoids recession, past relationships suggest that
equities will move higher from current levels, with the bond yield
curve flattening. Still, considerable uncertainty surrounds any
such forecasts, as both business cycles in general and the causes
of recessions in particular remain poorly understood. Hedge
assets thus retain considerable importance in asset allocation.
Having suffered a second downturn in 2010–12 following the
financial crisis recession, both the UK and the euro area find
themselves much earlier in their current expansions than the
U.S. In the former’s case, though, the rapid fall in joblessness
suggests a dynamic closer to that of the U.S. than to continental Europe, where the expansion is just beginning and seems
unlikely to run into capacity constraints for an extended period.
Business cycle history: The trend toward
lengthening
The U.S. economy has moved toward longer business cycles
over time, with more lasting expansions and briefer recessions.
This section defines the frameworks used to identify
expansions and recessions, describes the factors behind the
lengthening of the U.S. business cycle and provides guideposts
for assessing an economy’s cyclical condition.
The NBER thus looks for persistent, large-scale co-movement
in economic indicators, abstracting from temporary fluctuations in any one series. Troughs mark the beginning of expansions and peaks their end.
The NBER approach, which involves considerable discretion,
differs from commonly used journalistic rules of thumb, the
most prominent of which equates recessions with two consecutive quarters of negative gross domestic product (GDP)
growth. Along with GDP (admittedly the most important
barometer of activity), the NBER also looks at gross domestic
income (or GDI, which is the counterpart to GDP that tracks
income flows in the economy rather than output), as well as
employment and other monthly indicators such as retail sales
and industrial production. In this system, two back-to-back
negative quarters for GDP might not constitute a recession.
For example, a temporary collapse in one kind of activity (say,
mining) could push GDP growth below zero for six months,
while, at the same time, other parts of the economy are doing
fine. On the other hand, not every recession will involve two
consecutive down quarters. In the mild 2001 episode, for
example, the NBER identified a recession between March and
November. GDP fell in the first and third quarters of that year
but rose in the second. With the help of higher-frequency indicators than GDP, the NBER dating system also picks specific
months for peaks and troughs. Thus, the dating committee
believes that the current expansion began in June 2009, with
the previous one having ended in December 2007.
According to the NBER, since the mid-1950s (by which time the
disruptions of World War II had passed), the U.S. economy has
experienced 10 expansions.2 Exhibit 1 displays basic characteristics of these cycles.
U.S. postwar expansions have varied in length as well as magnitude
Useful understanding of business cycles and their relationship
with financial markets begins with a definition of terms. In the
U.S., most observers use the framework developed by the
National Bureau of Economic Research (NBER), which maintains a committee that dates business cycles by identifying
inflection points in economic activity. In its parlance, a recession entails “a significant decline in economic activity” that is
widespread, not confined to just a few sectors, and “can last
from a few months to more than a year.” Conversely, “during
an expansion, economic activity rises substantially, spreads
across the economy and usually lasts for several years.”1
“Statement of the NBER Business Cycle Dating Committee on the
Determination of the Dates of Turning Points in the U.S. Economy,”
www.nber.org/cycles/general_statement.html.
1
2 | Tracking the business cycle
EXHIBIT 1: U.S. EXPANSIONS SINCE 1954
Start
May ’54
Finish
Apr ’58
Feb ’61
Nov ’70
Mar ’75
Jul ’80
Nov ’82
Mar ’91
Nov ’01
Aug ’57
Apr ’60
Dec ’69
Nov ’73
Jan ’80
Jul ’81
Jul ’90
Mar ’01
Dec ’07
Jun ’09
Jun ’14
Duration
(in months)
39
24
106
36
58
12
92
120
73
60
Average GDP
Subsequent
growth rate peak-to-trough
(%)
decline (%)
3.8
-3.6
6.1
5.0
5.1
4.3
4.4
4.2
3.8
2.7
2.3
-1.3
-0.6
-3.1
-2.2
-2.6
-1.3
-0.3
-4.3
Source: National Bureau of Economic Research; data as of July 2014.
2
“U.S. Business Cycle Expansions and Contractions,” www.nber.org/cycles/
US_Business_Cycle_Expansions_and_Contractions_20120423.pdf.
The NBER has established dates of business cycles back to
1854, though in comparison with the decisions of the past
several decades, its judgments about the earlier periods should
be regarded as indicative. National accounts data do not
predate World War II, and other economic figures from the
prewar era may not be reliable. The NBER relied in part on
financial market behavior in making recession and expansion
determinations for the 1800s, in particular. As a group, though,
the earlier expansion and recession dates provide some flavor
for the economy’s cyclical performance before World War II in
comparison with more recent experience (Exhibit 2).
After its postwar recovery, Western Europe’s business cycles
have roughly coincided with those in the U.S.
EXHIBIT 3A: BUSINESS CYCLES IN THE UK
Start
Aug ’52
Finish
Aug ’75
May ’81
Mar ’92
Jan ’10
Sep ’74
Jun ’79
May ’90
May ’08
Aug ’10
Feb ’12
Jun ’14
Duration
(in months)
265
Average GDP
Subsequent
growth rate peak-to-trough
(%)
decline (%)
3.3
-2.9
46
108
194
7
3.7
3.7
3.3
1.7
2.2
28
-5.9
-2.4
-7.2
-0.6
EXHIBIT 3B: BUSINESS CYCLES IN GERMANY
Expansions have been generally longer than contractions
EXHIBIT 2: U.S. EXPANSIONS AND CONTRACTIONS SINCE 1854 (MONTHS)
150
Expansions
Contractions
Percent
100
50
Start
May ’67
Finish
Jul ’75
Oct ’82
Apr ’94
Aug ’03
Aug ’73
Jan ’80
Jan ’91
Jan ’01
Apr ’08
Jan ’09
Apr ’14
Duration
(in months)
75
54
99
81
56
63
Average GDP
Subsequent
growth rate peak-to-trough
(%)
decline (%)
5.0
-2.4
3.8
3.6
1.9
2.3
2.1
-2.7
-1.6
-0.8
-6.8
Source: Economic Cycle Research Institute; data as of July 2014.
0
-50
-100
1
3
5
7
9
11
13
15
17
19 21
23 25 27 29 31
33
Source: National Bureau of Economic Research; data as of July 2014.
The experience in Western Europe has broadly mirrored that of
the U.S. in recent decades, with one significant difference.
These economies suffered large-scale damage during World
War II that set them back considerably relative to the U.S.
Postwar reconstruction and renewed catch-up facilitated a
period of essentially uninterrupted growth into the 1960s and
1970s. Since that time, business cycles in Western Europe have
roughly coincided with their U.S. counterparts, with some
exceptions (the UK dodged the 2001 recession and thus
combined the U.S. expansions of the 1990s and early 2000s,
while parts of the euro area and the UK suffered a recession in
the 2010–12 period that interrupted the recovery from the
global financial crisis). Exhibits 3A and 3B show recent
business-cycle developments in the UK and Germany, as
identified by the Economic Cycle Research Institute (ECRI), a
research organization that follows a similar approach to the
NBER and tracks various global economies. The ECRI’s business
cycle dates for Japan, meanwhile, display that country’s unique
pattern, which combines the postwar reconstruction and
economic development megacycle experienced in Western
Europe with frequent recessions in more recent years, when
the economy has faced deflation and demographic headwinds.
The U.S. economy shows an unmistakable tendency toward
longer business cycles over time, as well as briefer (though
not necessarily less severe) recessions. Indeed, the most
recent three expansions (excluding the unfinished current
cycle) all stand among the five longest since 1854, joined by
the 1960s cycle and the boom fueled by public spending during World War II. Between 1854 and 1938, the economy spent
55% of the time (measured in months) in expansions, and
these episodes lasted 26 months on average. Those figures
rose to 82% and 49 months, respectively, in the 1938–1982
period and have climbed further to 91% and 86 months since
then. This final time span became known as the “Great
Moderation” before the global financial crisis struck. Although
the subsequent downturn proved particularly intense, many
aspects of the previous low-volatility environment have since
reasserted themselves.
J.P. Morgan Asset Management | 3 MARKET
INSIGHTS
PORTFOLIO
DISCUSSION:
Title Copy Here
Tracking
the
business cycle
What accounts for the lengthening of U.S. business cycles?
No consensus exists on the causes, but at least four factors
very likely have played a role:3
• Monetary policy formation has become significantly more
sophisticated in the past several decades (especially compared, obviously, with the 1800s, before the Federal Reserve
existed) and likely has served to dampen the economy’s
cyclical fluctuations. Heightened prudential regulation of the
financial sector likely has allied with purer interest-rate and
money-supply management to produce this effect.
• Manufacturing (where inventory cycles drive large swings in
activity) and agriculture (subject to weather-related gyrations) have declined in importance relative to the less volatile services sector.
• The federal government has grown larger, and its spending
typically follows a more stable path than private-sector
expenditure, in part because of explicit countercyclical
efforts (such as unemployment insurance and food stamps)
that cushion downturns and fade during booms.
• Broader availability of consumer credit appears to have
allowed households to smooth their own spending over life
cycles, making consumption less volatile than when its path
more closely tracked income (itself, in the past, often dependent on farm prices and other highly variable sources).
A more temporary circumstance—the rise of globalization—may
also have helped extend recent business cycles by leaning
against inflation and steadily boosting corporate profit margins
in the past 30 years or so. These phenomena allowed the
Federal Reserve to keep interest rates persistently low and
prevented significant downturns in business investment
spending (while perhaps also sowing the seeds of the virulent
financial crisis). With the globalization process largely complete,
this factor will likely play little or no role in future cycles, but
the other developments appear structural in nature. Thus,
longer business cycles likely represent a lasting phenomenon.
Stylized facts of business cycles
Business cycle patterns since the 1950s display a few persistent
features that may help in understanding the current position of
the economy. At the same time, other phenomena intuitively connected with cyclical fluctuations, like inflation, have in fact varied
widely across cycles, making them unsatisfactory guides to present conditions. The following “stylized facts” of business cycles can
serve as a guide for investors attempting to diagnose the economy’s cyclical condition. For data consistency reasons, the discussion focuses on the U.S., but the conclusions apply more broadly.
Cyclicality in business investment, housing and
durable goods purchases
These three types of expenditure help to define business
cycles, typically moving higher as expansions unfold and then
falling sharply during recessions.4 Exhibits 4–6 (next page)
show the behavior of nonresidential fixed investment (essentially business capital spending), residential construction and
consumption of durables, respectively, all measured as shares
of GDP (so that a rise implies faster growth than for the economy as a whole). In the exhibits, each line represents one
expansion, beginning in its first quarter (the trough) and continuing through its end (the peak). The legend identifies each
expansion by its starting quarter.
Among the three categories, nonresidential fixed investment
(or business investment spending) displays the clearest cyclical pattern. In nearly every expansion, it has risen almost
uninterruptedly as a share of GDP until the very end of the
expansion. Two exceptions stand out. In the 1980s, investment
peaked less than halfway through the cycle and declined
thereafter. Investment had begun that cycle at an unusually
elevated point, however. In addition, information technology
(IT) investment began assuming greater importance during
that period, and prices for IT products fell in relative terms in
this era. In the 1960s, investment also peaked well before the
end of the cycle, although in that instance it did not significantly decline until recession hit.
Somewhat muddier pictures characterize the other two spending types. Residential construction has generally surged early
in expansions but has often cooled well before their ends. The
2000s expansion represents the most obvious example of this
kind of reversal, as the economy continued to grow for a while
3
J. Bradford DeLong and Lawrence H. Summers, “The Changing Cyclical
Variability of Economic Activity in the United States,” The National Bureau
of Economic Research, NBER Working Paper No. 1450, November 1986.
4 | Tracking the business cycle
4
Finn E. Kydland and Edward C. Prescott, “Business cycles: Real facts and a
monetary myth,” Federal Reserve Bank of Minneapolis, Quarterly Review,
Spring 1990.
Business investment spending claims a larger share of GDP in
U.S. business expansions
EXHIBIT 4: U.S. NONRESIDENTIAL FIXED INVESTMENT BY CYCLE
16
54Q1
15
58Q1
60Q4
70Q4
13
75Q1
12
80Q3
82Q3
11
91Q1
10
01Q3
9
09Q2
8
In every U.S. expansion since the 1950s, the unemployment
rate has fallen after a spike (of varying size) during the previous recession (Exhibit 7). Joblessness has typically dropped
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Quarters
Source: J.P. Morgan; data as of March 2014.
Residential construction has generally surged early in expansions
EXHIBIT 5: U.S. RESIDENTIAL CONSTRUCTION BY CYCLE
7
54Q1
58Q1
6
70Q4
% of GDP
Decline in U.S. unemployment depends on the length of
economic cycles
EXHIBIT 7: U.S. UNEMPLOYMENT RATE
12
Recession
60Q4
10
75Q1
5
80Q3
82Q3
4
91Q1
8
6
01Q3
3
2
Falling unemployment
Percent
% of GDP
14
after the housing bubble began to deflate (its popping, of
course, eventually helped lead to the 2008 recession). On several other occasions, though, housing weakened around the
mid-point of the cycle, a phenomenon that, in some cases,
likely reflected a first wave of monetary tightening. Durable
goods purchases have behaved somewhat similarly, with
strength in the first half of the cycle, followed by a leveling off.
As with business investment, durable goods softened around
the halfway point of the 1980s expansion and drifted lower as
a share of GDP until the 1990 recession.
09Q2
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Quarters
4
2
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
Source: J.P. Morgan; data as of March 2014.
Source: J.P. Morgan; data as of June 2014. Shading denotes recessions.
Durable goods purchases tend to hold steady until recession strikes
EXHIBIT 6: U.S. DURABLE GOODS CONSUMPTION BY CYCLE
11
54Q1
58Q1
10
60Q4
% of GDP
70Q4
75Q1
9
80Q3
82Q3
8
91Q1
01Q3
7
6
09Q2
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Quarters
steadily during expansions, so that the total reduction has
depended on the length of the cycle. In all but two cases,
recessions began with the unemployment rate at or below
5.5%, similar to the Federal Open Market Committee’s (FOMC)
current estimate of the neutral rate. The recessions that began
in January 1980 and July 1981 represent the exceptions. In the
first of these, an oil shock and a very large shift in the Federal
Reserve’s reaction function aimed at halting the lengthy rise in
inflation helped end the expansion, which had begun with very
high joblessness after the deep 1973–75 downturn, while the
second also involved the Federal Reserve’s war against inflation and followed the briefest expansion of the postwar era.
Source: Haver Analytics, J.P. Morgan; data as of March 2014.
J.P. Morgan Asset Management | 5 MARKET
INSIGHTS
PORTFOLIO
DISCUSSION:
Title Copy Here
Tracking
the
business cycle
Inflation in recent decades has not displayed much short-term
cyclicality
EXHIBIT 8: U.S. CORE CONSUMER PRICES BY CYCLE
14
May-54
12
Apr-58
Feb-61
% Y-o-Y
10
Nov-70
8
Mar-75
6
Jul-80
Nov-82
4
Mar-91
Nov-01
2
Jun-09
0
-2
1
13
25
37
49
61
73
Months
85
97
109
121
Source: J.P. Morgan; data as of May 2014.
Lack of credit and saving cyclicality
Bank lending and its cousin, the household saving rate, also
have not displayed consistent cyclical patterns. While the saving rate fell during the course of the late 1970s, 1980s and
1990s expansions, in the broader context, those declines
appear to form part of a more secular long-term downward
trend (Exhibits 9 and 10). Credit, for its part, has grown a bit
faster than GDP fairly consistently since the 1950s, without an
obvious link to the business cycle (Exhibit 11, next page). The
2000s expansion represents an exception, with a strong credit
boom that helped sow the seeds of the subsequent recession.
The saving rate fell during the late 1970s, 1980s and 1990s
expansions
EXHIBIT 9: U.S. PERSONAL SAVING RATE BY CYCLE
18
Lack of inflation cyclicality
What accounts for this result? Essentially, rather than moving
in lockstep with growth, inflation in the U.S.—and in most
other developed economies—has gone through one large cycle
in the postwar era, rising in the 1960s and 1970s and falling
since. Changing central bank behavior and the resulting
lengthy de-anchoring and then re-anchoring of inflation
expectations drove much of that increase and decline (along
with other phenomena, like the productivity growth slowdown
of the 1970s and the globalization of the 1990s and 2000s).
With central banks satisfied with their efforts at squeezing
inflation out of the system and now, in many cases, attempting
to push it up from undesirably low levels, inflation may reattach itself to the business cycle in coming years. Such a connection, though, would mark a break from the experience of
the past 40 years.
6 | Tracking the business cycle
Nov-70
14
Mar-75
Jul-80
12
Nov-82
% Y-o-Y
10
Mar-91
8
Nov-01
6
Jun-09
4
2
0
1
13
25
37
49
61
73
Months
85
97
109
121
Source: J.P. Morgan; data as of May 2014.
The saving rate declined secularly from 1973 onward, but jumped
in recessions
EXHIBIT 10: U.S. PERSONAL SAVING RATE
20
Recession
15
Percent
Whereas an abstract conception of a business cycle would
likely involve rising inflation during the expansion, reality in
recent decades has proven quite different. Since at least the
1970s, inflation has shown almost no cyclicality. Exhibit 8
shows monthly core consumer price inflation, pegged to each
expansion, with starting months identified in the legend
(overall inflation is used for the May 1954 expansion, as core
data go back only to 1958). Strikingly, in each of the past six
expansions, core inflation (measured in year-on-year terms
and using a three-month average to smooth out temporary
fluctuations) has stood at a lower level in the final month than
in the starting month. In other words, inflation by itself
appears to have played a very small role in driving business
cycle fluctuations.
Feb-61
16
10
5
0
1959 1964 1969 1974 1979 1984 1989 1994 1999 2004 2009 2014
Source: J.P. Morgan; data as of May 2014.
EXHIBIT 11: U.S. BANK CREDIT BY CYCLE
70
54Q1
58Q1
60Q4
% of GDP
60
70Q4
75Q1
80Q3
50
82Q3
91Q1
01Q3
40
30
09Q2
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Quarters
Source: J.P. Morgan; data as of March 2014.
Productivity up, down, then sometimes up again
The growth rate of (labor) productivity in the economy typically
surges in the first year or so of an expansion before slowing
again (Exhibit 12). This pattern likely reflects typical business
behavior, as employers wait for some time after demand
recovers to add workers and instead boost their output with
existing workforces, thus raising productivity, before growing
confident enough to expand staffing. In longer-lived expansions, productivity growth then appears to reaccelerate. This
renewed pickup may reflect favorable technological shocks (as
may have been the case in the 1990s expansion), demand
shocks (for example, fiscal expansion related to the Vietnam
War in the 1960s) or other phenomena, and may not be susceptible to a single explanation across cycles.
Labor productivity has surged in the first year of an expansion as
industry takes advantage of unused capacity
Policy interest rates generally up
As is well known, policy interest rates tend to rise during
expansions. As with most of the other phenomena previously
discussed, however, no hard and fast rule has applied in this
area. The Federal Reserve’s framework has changed over time,
with the funds rate (as distinct from the money supply or, in
an earlier period, the discount rate) playing an increasing role
in defining the policy stance since the 1980s. The 1990s and
2000s cycles mirrored each other in featuring ongoing rate
cuts in the early stage of the expansion and a round of rate
hikes about three years into the cycle (Exhibit 13). In neither
case did rates rise significantly further as the expansion
matured. In several earlier cycles, Federal Reserve policy did
tighten at the end of the cycle, pushing up short-term rates.
More recently, however, the Federal Reserve does not appear
to have systematically administered coups de grâce that
pushed the economy into recession.
The Fed generally tightens, but rate hikes do not appear to have
ended most expansions
EXHIBIT 13: U.S. FEDERAL FUNDS RATE BY CYCLE
25
May-54
Apr-58
20
Feb-61
Nov-70
Percent
Credit growth has outpaced and helped fuel U.S. expansions since
the 1950s
Mar-75
15
Jul-80
Nov-82
10
Mar-91
Nov-01
5
0
Jun-09
1
13
25
37
49
61
73
Months
85
97
109
121
Source: J.P. Morgan; data as of July 2014.
EXHIBIT 12: U.S. PRODUCTIVITY BY CYCLE (NONFARM BUSINESS)
8
54Q1
% Y-o-Y
7
58Q1
6
60Q4
5
70Q4
4
75Q1
80Q3
3
82Q3
2
91Q1
1
01Q3
0
09Q2
-1
-2
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
Quarters
What ends business cycles?
Business cycle dynamics, particularly recession drivers,
remain poorly understood. This section starts by examining
the theoretical attempts to identify predictable patterns,
while evaluating the current economic cycle for potential
clues to forecasting the next downturn. The verdict: The preponderance of evidence so far suggests the U.S. expansion
remains in the early-to-middle stage, while cycles in the UK
and the euro area seem even less advanced.
Source: J.P. Morgan; data as of March 2014.
J.P. Morgan Asset Management | 7 MARKET
INSIGHTS
PORTFOLIO
DISCUSSION:
Title Copy Here
Tracking
the
business cycle
As the discussion of stylized facts suggests, considerable variation has characterized even the fairly small number of expansions in the U.S. during the postwar era. Business cycles, thus,
have not proven susceptible to simple or uniform explanation.
In part, definitional ambiguities and data shortfalls complicate
the task of testing hypotheses: “Business cycles involve numerous activities and are not adequately represented by specific
cycles in any single variable…no comprehensive time series
exist to cover their long and varied history.”5 Even beyond
these statistical issues, limited commonality across expansions
has frustrated attempts to hypothesize about their underlying
drivers and, equally importantly, the causes of recessions.
In an attempt to identify predictable patterns that might assist
in forecasting downturns, economists have conducted a great
deal of statistical analysis in search of periodicity in business
cycles.6 For example, the economy might follow a “limit cycle”
with a floor and ceiling. Investment spending would typically
serve as the main driver of the business cycle process in such
a model, which could include random shocks to prevent strictly periodic outcomes. These approaches have not borne significant fruit, unsurprisingly, in light of the earlier description of
business cycle history, with its widely varying outcomes and
tendency toward longer expansions over time.
At the same time, statistical tests have also run into difficulty
attempting to determine, on the one hand, whether expansions
are random walks (so that the economy faces a basically
constant probability of recession in any given time period) or,
alternatively, whether the probability of recession rises
gradually over time (so that expansions are killed simply by an
aging process). The random walk hypothesis clashes with the
fact that few business cycles end after very short intervals, as
well as with various regularities that do exist across expansions,
both in the U.S. and elsewhere. But the “old-age” view also
comes up short, again in part because of the observed
tendency toward longer expansions, which presumably reflects
some change in genuine underlying drivers: “What matters
primarily…is not the passage of calendar time but what happens
over time in and to the economy in question.…Knowledge of the
current phase of the business cycle and its age can help but
must not be used in isolation.”7
5
6
7
Victor Zarnowitz, Business Cycles: Theory, History, Indicators and Forecasting
(The University of Chicago Press, 1992): p. 237.
Zarnowitz (1992), pp. 251–264 for an extensive discussion of these
approaches.
Zarnowitz (1992), p. 262.
8 | Tracking the business cycle
A related statistical approach with a stronger theoretical overlay attributes business cycle periodicity to political events. In
this framework, governments attempt to boost their survival
chances by stimulating booms just ahead of elections, with
payback in the form of recessions early in the next term (to get
the downturn out of the way before voting time comes around
again). This so-called political business cycle theory has lost
academic adherents over time but still holds some sway, particularly among financial market analysts. Its foundations, though,
also appear weak. Even in an earlier era of unified (or at least
less polarized) government, the idea that policymakers could
sufficiently fine-tune the economy so that its movements would
synchronize perfectly with the electoral calendar strained credulity, as did the notion that the public would accept such persistent behavior. In any case, under today’s political circumstances, such an approach to policymaking seems impossible.
Using theoretical frameworks to explain
business cycles
Models with stronger theoretical underpinnings, especially
those that attempt to ground macroeconomic developments in
microeconomic descriptions of behavior, tend to fall into two
broad groups. The first, which ultimately derives from
Keynesian thinking, explains recessions in part as periods of
demand shortfall that reflect the failure of markets to clear,
perhaps because of incomplete price adjustments. According
to this line of argument, recessions reflect negative shocks
either to consumption or investment or both. Investment
cycles could owe to a variety of causes, including uncertainty,
interest rates or other sudden increases in the cost of capital,
with some role for a declining marginal efficiency of capital as
expansions unfold. Consumption shocks might stem from foresight of declining income or from inter-temporal preference
shifts (or, for that matter, panics).8,9
The second group of models, which descends from neoclassical
views of the economy, assumes that markets clear continuously,
leaving no role for disequilibria between demand and supply.
Instead, these “real business cycle” frameworks explain
downturns primarily as the consequence of technology shocks
(“technology” in this context should be understood as anything
that affects total factor productivity, or the portion of
fluctuations in economic output that do not stem purely from
8
9
Olivier Blanchard, “Consumption and the Recession of 1990-91,”
MIT Working Paper No. 93-5.
John T. Harvey, “The US business cycle since 1950: A post Keynesian
explanation,” Texas Christian University, Working Paper No. 10-05,
August 2010.
changes in labor and capital stock inputs). One limitation of this
approach is its ambiguity about what exactly would constitute
such a shock, as presumably the disappearance of existing
technology lacks plausibility. One line of argument emphasizes
regulatory change, while another leans on increases in
uncertainty that might reflect concerns about policy or other
structural aspects of the environment.10,11,12
A cycle-by-cycle look at expansions and their ensuing recessions
underscores the multiplicity of drivers.13 The expansions that
began in November 1970 and March 1975, for example, likely
came to their ends largely because of spikes in the oil price,
which otherwise appears to have played only a limited role (at
least in starting or stopping expansions). The April 1958 and
July 1980 expansions, on the other hand, seem to have ended
because of shifts in the Federal Reserve’s approach to monetary
policy (such that the downturn that began in July 1981 is often
combined with the previous contraction into a single phenomenon
known as the “Volcker Recession,” with the Federal Reserve significantly changing its approach in order to squeeze inflation
out of the system). While interest rates have risen during the
course of some other cycles (though not all), on those occasions
tighter monetary policy appears to have served as an endogenous force—having responded to broader developments in the
economy—and in that sense did not represent the cause of subsequent recessions. On two occasions—at the end of the May
1954 and the long February 1961 expansions—fiscal policy tightened significantly, partly because of a sharp scaling back of military spending. The November 2001 expansion, of course, ended
with a financial crisis. The end of the November 1982 expansion
lacks a compelling explanation, although the invasion of Kuwait
and the resulting damage to consumer confidence (along with
higher oil prices) may have played a significant part. Real business cycle theorists attribute the recession to a technology
shock (but perhaps one related to the uncertainty generated by
military conflict).
Research about the UK, where business cycles have perhaps
been the most studied outside of the U.S., has turned up similar ambiguities. The smaller number of cycles in the UK during
the postwar era complicates analysis by reducing sample size.
10
11
12
13
Gary D. Hansen and Edward C. Prescott, “Did technology shocks cause the
1990-1991 recession?” The American Economic Review, Volume 83, Issue 2,
May 1993.
Nicholas Bloom, et al. “Really uncertain business cycles,” NBER. Working
Paper No. 18245, July 2012.
Alejandro Justiniano et al. (2009), “Investment Shocks and Business
Cycles,” Journal of Monetary Economics, Volume 57, Issue 2, March 2010.
Peter Temin, “The causes of American business cycles: An essay in
economic historiography,” NBER Working Paper No. 6692, August 1998.
Meanwhile, expansion end dates in the UK have broadly, if
imperfectly, mirrored the U.S. experience, suggesting a similar
array of causes, including oil and monetary policy shocks,
along with the recent addition of the euro area crisis.
Fluctuations in investment spending appear to lie at the heart
of UK cycles, similar to the U.S., but whether or not monetary
policy has played a significant role in driving either shocks to
investment or other cyclical dynamics is a matter of controversy. One significant piece of research done partly at the Bank of
England concluded that “a large component of GDP fluctuations are unpredictable and not Granger (i.e., statistically)
caused by a standard list of macroeconomic variables.”14,15
Evaluating the current cycle
Given the large variation in historical business cycle experience and still-limited theoretical understanding of what drives
expansions and recessions, any discussion of current conditions must start with an admission that no definitive statements are possible. Judgment will inevitably play a major role
in evaluating business cycle conditions. At the moment,
though, a preponderance of evidence suggests that the U.S.
expansion remains in its early-to-middle stage. Cycles in the
UK and especially in the euro area seem even less advanced.
These economies thus appear to possess considerable
“room to run” before recession risk becomes elevated.
Expansion age does not seem a concern
The U.S. expansion has just passed its fifth birthday, making it
already reasonably old by long-term standards. But the most
recent three expansions, including the November 2001 cycle
that was cut short by the financial crisis, averaged 95 months
in length, and the trend appears to point toward lengthening.
Moreover, as discussed earlier, statistical tests point to only a
weak relationship between an expansion’s duration and the
likelihood of recession. Thus, the length of the current expansion in the U.S. does not appear inherently problematic. In the
euro area, the region as a whole emerged from recession only
last year, although Germany has been growing continuously
since the start of 2009. The UK, for its part, began its expansion in early 2012. Thus, Europe finds itself in the early stages
of the present cycle.
14
15
Allison Holland and Andrew Scott, “The determinants of UK business
cycles,” Bank of England working paper number 58, January 1997.
Juan Paez-Farrell (2003), “Business cycles in the United Kingdom: An
update on the stylized facts.” The University of Hull Business School,
Research Memorandum, July 2003.
J.P. Morgan Asset Management | 9 MARKET
INSIGHTS
PORTFOLIO
DISCUSSION:
Title Copy Here
Tracking
the
business cycle
Weak growth thus far
A sluggish leverage cycle
Relative to past postwar cycles, the current expansion stands
out both for the sluggishness of growth and the depth of the
preceding recession. As a result, the level of real GDP stands
just 6% above its pre-recession peak (Exhibit 14). Only one
recession since the 1950s has occurred with GDP so close to
its previous high. That incident—the 1981-82 downturn—happened after just five quarters of growth and resulted from the
structural shift in Federal Reserve policy that took place at
that time (indeed, this dip is often regarded as the second half
of the aforementioned “Volcker Recession,” the first leg of
which occurred in the first half of 1980). This analysis (and the
exhibit) does not take into account changes in the economy’s
potential growth rate over time. Slower-trend growth today,
compared with previous decades, means that the current
expansion should not be held to earlier standards. Nonetheless, the level of GDP at the moment does not suggest a
high risk of overheating.
A combination of regulatory change and chastened borrowers
means that bank credit has barely expanded in the past few
years. Total credit-related U.S. bank assets rose 2.1% yearover-year in the first quarter of 2014. By contrast, during the
comparable quarter of the 2000s cycle, bank credit rose at
nearly a double-digit clip, and even in the more restrained
1990s expansion, credit was climbing at a 7% pace by the fiveyear point. In the euro area, bank lending is still shrinking
(Exhibit 15), while in the UK, credit to corporations is dropping
steadily, while lending to households is rising at less than 2%
year-over-year. Indeed, reduced ability to take leverage
appears to mark a significant contrast between the current
expansion and past cycles, limiting the ability of imbalances to
form this time around.
Euro area lending is still feeling the effects of the financial crisis
EXHIBIT 15: EURO AREA BANK CREDIT BY TYPE
20
Household
Corporate
The current economic expansion is weaker than past expansions
EXHIBIT 14: U.S. REAL GDP BY CYCLE (PRE-RECESSION PEAK = 100)
54Q1
150
58Q1
140
70Q4
60Q4
10
5
75Q1
130
Index
80Q3
120
82Q3
110
01Q3
91Q1
09Q2
100
90
15
% Y-o-Y
160
0
-5
2004
2006
2008
2010
2012
2014
Source: J.P. Morgan; data as of May 2014.
1
5
9
13
17
25
21
Quarters
29
33
37
41
Source: J.P. Morgan; data as of March 2014.
Low interest-sensitive spending
Considering that the most consistent business-cycle regularity
is a rise in big-ticket, credit-sensitive spending, the behavior of
these items in the U.S. over the past few years provides reassurance about the sustainability of the current expansion.
Business investment spending has broadly followed the pattern of previous expansions and does not appear to have
reached a worrisomely high level. Meanwhile, both housing
investment and consumption of durables—each of which
reflects household decisions—are running at extremely low
levels by long-term standards and have risen only modestly
from their troughs.
10 | Tracking the business cycle
A reasonable household saving rate
After jumping during the recession, the U.S. household saving
rate has moved downward in the past few years. Still, thus far
in 2014, it has averaged a full percentage point above its level
during the corresponding period of the 2000s expansion.
Moreover, it does not look low relative to the current condition
of household balance sheets, with net worth as a share of disposable income near its all-time high. At this point, the likelihood of a sudden jump in the household saving rate (producing a period of significant weakness in consumer spending),
motivated by a desire to rebuild household net worth, appears
limited. Saving rates in the UK and Germany also look high
enough to provide some cushion against unexpected shocks to
household income in those countries.
At least since the 1980s, inflation has not moved distinctly in
line with business cycles, and recessions have struck even with
inflation moving lower. Still, with the long transition from high
to low inflation now complete, growth and inflation may
resume a more intuitive relationship, with price pressures
serving as a signal of broader capacity constraints and imbalances facing the economy. On that front, the still-low levels of
both wage and price inflation in the U.S. suggest ample slack.
Although inflation may have bottomed out in early 2014, it is
running below the Federal Reserve’s target, and expectations
appear well-anchored. The FOMC thus seems likely to tolerate
a fair degree of acceleration from current levels before feeling
the need to tighten policy to such a degree that recession risk
would rise significantly. The UK appears to find itself in a similar position, although the Bank of England probably enjoys
less room for maneuver than the Federal Reserve because of
its recent history of inflation overruns. By contrast, the
European Central Bank (ECB) has become increasingly concerned about excessively low inflation, another factor suggesting that the euro area in particular finds itself in the very early
stages of its expansion.
Labor market slack
Perhaps the most controversial and uncertain aspect of
business cycle evaluation in the U.S. at the moment concerns
labor market conditions. Joblessness has fallen rapidly and
stands less than a percentage point above the Federal Reserve’s
current estimate of a neutral long-term unemployment rate.
Yet the employment-to-population ratio, a broader measure of
labor market pressure, has risen only modestly from its
recession trough (Exhibit 16). One measure suggests that the
U.S. labor force participation lags even as unemployment has fallen
EXHIBIT 16: U.S. EMPLOYMENT/POPULATION RATIO AND
UNEMPLOYMENT RATE
66
Employment/population ratio
Percent
6
7
8
Percent
5
58
9
10
56
11
54
1960
1969
1978
1987
1996
2005
EXHIBIT 17: U.S. LABOR FORCE PARTICIPATION RATE AND
DEMOGRAPHIC TREND
67
Actual
Demo trend
66
65
64
62
2008
4
60
Aging population may not account for all of the drop in labor
force participation
2
3
62
Demographic drift, as more of the Baby Boomers reach retirement age, almost certainly explains some of the drop in the
labor force participation rate (the share of the working-age
population that is actively involved in the labor market), and
this aging process will not reverse. In other words, many of
the exits from the labor force observed in recent years are
permanent. But population aging alone appears to account for
only about half of the drop in participation since 2007, leaving
open the possibility that much of the remainder has reflected
more cyclical phenomena, with workers becoming discouraged
(Exhibit 17). As the job market firms, some of these sidelined
individuals will likely return, boosting the labor force growth
rate. Satisfactory growth could thus coexist with a slower
decline in the unemployment rate and with an ongoing moderate pace of wage and price inflation.
63
Unemployment rate
64
economy has already absorbed most of the slack created by the
downturn, while the other implies that this process has barely
begun. If the headline unemployment rate is sending the right
message, then wage inflation might be expected to accelerate
fairly soon, and the business cycle has likely reached a somewhat advanced stage. If, instead, the decline in joblessness
overstates the degree of improvement in the labor market, then
the economy likely can continue to grow for an extended period
before hitting supply constraints.
Percent
Low wage and price inflation
12
2014
2009
2010
2011
2012
2013
2014
Source: J.P. Morgan; data and projections as of June 2014. Demographic trend
fixes participation rates for each age cohort at the 2007 level, with
subsequent changes reflecting shifts in population by age cohort.
Resolution of this question will play a major role in determining
how the U.S. business cycle unfolds from here. If considerable
slack remains in the labor market, then the Federal Reserve can
afford to be patient and gradual in its removal of monetary policy
Source: J.P. Morgan; data as of June 2014.
J.P. Morgan Asset Management | 11 MARKET
INSIGHTS
PORTFOLIO
DISCUSSION:
Title Copy Here
Tracking
the
business cycle
support, and the expansion will likely run for several more
years. On the other hand, if labor market slack has been largely
reabsorbed already, then the Federal Reserve will likely need to
adopt a more vigilant stance soon, and the possibility of recession will rise. A downturn would not loom as an inevitability—
with inflation expectations well-anchored, a “high pressure”
economy need not necessarily generate rapid inflation, especially if productivity growth picks up as it did in the latter stages
of the 1990s expansion—but markets would need to price in
some risk of an end to the expansion. The path of the labor
force participation rate, thus, will occupy center stage in U.S.
business cycle analysis during the coming months and years.
On the labor market front, conditions differ between the euro
area and the UK. In the single currency zone, the two punishing recessions of the past six years have created enormous
slack in this area, with the headline unemployment rate nearly
12% and only marginally below its 2013 peak. Although the
euro area has struggled with long-term unemployment in the
past, the jobless rate averaged 8.7% between the euro’s inception (in 1999) and 2007, more than three percentage points
below the current level. In terms of labor market spare capacity,
the euro area stands convincingly in the early stages of its
expansion. The UK jobless rate, by contrast, has behaved
somewhat similarly to the U.S. unemployment rate, dropping
by 1.6 percentage points since the end of 2011. In the UK’s
case, though, this decline does not reflect falling labor supply,
as the participation rate has risen during that same period
(Exhibit 18). Instead, the sharp improvement owes to fairly
strong economic growth and a disproportionately rapid pace
of job gains relative to growth (i.e., sluggish productivity
increases). At 6.8%, the unemployment rate currently stands
The UK has not experienced the same discrepancy between the
unemployment rate and labor force participation
EXHIBIT 18: UK UNEMPLOYMENT AND LABOR MARKET ACTIVITY
RATES
Unemployment rate
Activity rate
11
64.5
10
64.0
9
63.5
8
63.0
7
62.5
6
62.0
5
61.5
4
1990
1993
1996
1999
2002
Source: J.P. Morgan; data as of April 2014.
12 | Tracking the business cycle
2005
2008
2011
2014
61.0
Little commodity stress
The oil price spikes of 1973 and the late 1970s, along with the
1990 jump, likely played significant roles in causing recessions
at those times. At this point, such an “out-of-the- blue” end to
the expansion appears unlikely for several reasons. First, the
stable behavior of commodity prices during the past few years
suggests very little stress in the system. Second, the wave of
commodity investment since the mid-2000s, although now
fading, appears to have flattened supply curves significantly.
Third, developed economies have become significantly less
commodity-intensive than was the case at the time of the oil
shocks. U.S. petroleum consumption in 2013, for example,
stood just 9% above that of 1973, despite real economic
growth of 191% during the same interval.
Thus, current conditions do little to signal an elevated probability of recession in the near term. The low level of interestsensitive spending, the most cyclical component of the economy, sends a particularly encouraging message on this front, as
does the limited amount of leverage-taking that has occurred
thus far. How long might the expansion run? No reliable forecast is possible, but the two expansions that preceded the
November 2001 cycle (which ended prematurely thanks to the
financial crisis) lasted an average of 106 months. Matching
that experience would suggest roughly another four years
before recession hits, although the unusually timid nature of
the present cycle implies a significant possibility of a more
stretched-out expansion that lasts longer.
A need for humility
65.0
Percent
Percent
12
only a bit above the Bank of England’s current 6.0-6.5% estimate of a medium-term equilibrium rate, although the Bank
does believe that the neutral rate itself may fall over time.16
In this respect, at least, the UK economy appears modestly
ahead of the U.S. in the current expansion, with the euro area
trailing far behind both.
The earlier discussions of business cycle history and theory
illustrated the limits of current understanding in this area.
Indeed, economists have generally not succeeded very well in
predicting recessions. The financial crisis and subsequent turmoil in the euro area demonstrated this point. One study
showed that forecasts collected by Consensus Economics in
September 2007 did not anticipate any of the 62 recessions
that occurred globally in 2008-09 (the vast majority of which,
of course, were related to the financial meltdown, which few
16
Bank of England, Inflation Report, May 2014, pp. 28–30.
The strength of stock rallies has varied, but the rallies have
tended to last across the entire length of economic expansions in
the U.S.
EXHIBIT 19: S&P 500 BY CYCLE (FIRST MONTH OF EXPANSION = 100)
450
Apr-58
350
Feb-61
The strongest regularity between cycles and markets occurs in
U.S. large-cap equities, which tend to move higher throughout
expansions, peaking shortly before recession hits. Exhibit 19
illustrates this link. Significant corrections in the middle of
expansions have occurred only rarely, with the 1987 stock
market crash the major exception. Even in that instance, equities stabilized soon and eventually made new highs before the
expansion ended. The 1990s equity bubble, for its part, ended
only as recession approached, with the market peaking seven
months before the expansion ended. In the current period,
equities are broadly mirroring their previous cyclical behavior,
with particular resemblance to the extended 1982 and 1991
17
Hites Ahir and Prakash Loungani, “Fail again? Fail better? Forecasts by
economists during the great recession,” presentation at George Washington
Research Program in Forecasting Seminar, January 30, 2014.
Jul-80
200
Nov-82
150
Mar-91
Nov-01
100
Jun-09
50
1
13
25
37
49
61
73
Months
85
97
109
121
Source: J.P. Morgan; data as of June 2014.
expansions. Those experiences suggest little constraint on further equity gains as long as the economy avoids recession.
UK equities have followed a less consistent pattern. Although
they, too, have tended to move higher during expansions, they
have not done so uniformly (Exhibit 20). In 1987, for example,
UK stocks plunged alongside their U.S. counterparts and only
marginally topped their pre-crash peak in the remainder of the
expansion, even though the economy continued to grow for
nearly three additional years. Similarly, in the long postwar
expansion, equities fell sharply in 1969-70 and did not recover
until late 1971. An even more extended funk took place between
2000 and 2003, with stocks topping their 1999 level only in
2006. Expansions in the UK thus have not provided reliable
insurance against equity slumps. This behavior likely reflects the
Stock rallies in the UK don’t have the same consistent
relationship with periods of economic growth as in the U.S.
EXHIBIT 20: UK EQUITIES BY CYCLE (FTSE ALL-SHARE, FIRST MONTH OF
EXPANSION = 100)
400
Aug-52
350
Aug-75
May-81
300
Mar-92
250
Jan-10
Index
Equities rise throughout expansions
Mar-75
250
0
While expansions themselves have displayed great variety over
time, the connection between cycles and financial markets has
shown several significant and fairly consistent patterns. The
economy’s business cycle position has played a dominant role
in driving capital market outcomes. Thus, as a core component
of their asset allocation process, investors need to take views,
however provisional, about where the economy stands in the
cycle and the probability of near-term recession.
Nov-70
300
Markets and cycles
Economic cycles have played a dominant role in driving
capital market outcomes. As a result, investors need to
consider where the economy stands in the cycle and the
probability of near-term recessions as factors in their asset
allocation process. This section looks at the investment
implications and explains why a tilt toward growth-sensitive
assets is still warranted.
May-54
400
Index
analysts foresaw). Most of the 19 recessions that happened in
2011–12 (primarily as a result of the euro area panic) also went
unanticipated.17 While some of this track record could reflect
the lack of incentives for private-sector forecasters to predict
recessions, the more likely culprits are unreliable and incomplete models, as well as shocks that are difficult to anticipate.
This being the case, any evaluation of current conditions carries a wide confidence interval. Although an imminent recession appears unlikely, it is still a possibility.
Feb-12
200
150
100
50
0
1
25
49
73
97
Months
121
145
169
193
Source: Haver Analytics, J.P. Morgan; data as of May 2014.
J.P. Morgan Asset Management | 13 MARKET
INSIGHTS
PORTFOLIO
DISCUSSION:
Title Copy Here
Tracking
the
business cycle
Credit spreads tighten most at the onset of the growth cycle…
…and widen as the expansion matures
EXHIBIT 21A: U.S. INVESTMENT GRADE CREDIT SPREAD OVER
TREASURIES BY CYCLE
EXHIBIT 21B: U.S. HIGH YIELD CREDIT SPREAD OVER TREASURIES
BY CYCLE
2.5
Nov-01
2.0
0.5
0.0
0.5
0.0
-0.5
-1.0
-1.0
1
13
25
37
49
61
73
Months
85
97
109
Jun-09
1.0
-0.5
-1.5
Nov-01
1.5
Basis points
1.0
Mar-91
2.0
Jun-09
1.5
Basis points
2.5
Mar-91
-1.5
121
1
13
25
37
49
61
73
Months
85
97
109
121
Source: Bloomberg, J.P. Morgan; data as of June 2014.
Source: Bloomberg, J.P. Morgan; data as of June 2014.
long nature of UK business cycles, during which the economy
has sometimes experienced significant headwinds that have not
quite tipped it into recession; the relatively open nature of the
economy and the large foreign operations of UK companies,
which have exposed them to earnings shocks even when the
local economy has not felt distress; and the dominance of the
U.S. in global equity trading, such that other markets tend to
follow developments in the S&P 500 (as in 1997).
occurs fairly early on in the expansion. Third, in contrast with
equities, spreads can widen significantly while the expansion is
under way. This last observation relates to the 1990s cycle,
when spreads troughed nearly four years before recession hit
and moved noticeably wider thereafter. This episode appears
related to the Asian financial crisis and the subsequent failure
of Long-Term Capital Management, both of which generated
significant volatility that spurred a cutting back of risk-taking
among broker-dealers and high-frequency traders. The overthe-counter nature of bond trading may make this asset class
more vulnerable to financial shocks than is the case for
exchange-traded equities. In the 2000s cycle, by contrast,
credit spreads widened only when the recession was imminent
(when the financial crisis began unfolding). Based on this
rough template, the outlook for credit spreads, on the
assumption that neither recession nor a financial shock is
Credit spreads fall during expansions, but can
trough before recession hits
Reliable data for credit spreads do not go back very far, lending uncertainty to conclusions. The behavior of credit spreads
in the past two cycles, though, tentatively suggests three patterns (Exhibits 21A and 21B). First, spreads tend to move
lower during expansions. Second, most of their tightening
Nominal risk-free yields have not shown a consistent pattern…
...as is also the case with real rates
EXHIBIT 22A: U.S. TREASURY YIELDS BY CYCLE
EXHIBIT 22B: U.S. REAL TREASURY YIELDS BY CYCLE
16
May-54
14
Apr-58
10
Nov-70
8
Mar-75
6
Jul-80
4
Nov-82
Mar-91
2
Nov-01
0
Jun-09
-2
1
13
25
37
49
61
73
Months
Source: J.P. Morgan; data as of June 2014.
14 | Tracking the business cycle
85
97
109
121
May-54
Apr-58
8
Feb-61
6
Percent
Feb-61
Percent
12
-4
10
Nov-70
Mar-75
4
Jul-80
2
Nov-82
Mar-91
0
Nov-01
Jun-09
-2
-4
1
13
25
37
49
61
73
Months
Source: J.P. Morgan; data as of June 2014.
85
97
109
121
imminent, likely includes some additional tightening, although
most of the move from their wide point during the recession
has probably already taken place.
No consistent trend in Treasury yield levels, but
a tendency toward flattening
Longer-term risk-free rates have not displayed an obvious
cyclical pattern, especially in the past few decades. In nominal
terms, 10-year U.S. Treasury yields have broadly reflected the
postwar inflation cycle, generally rising through the course of
cycles between the 1950s and the early 1980s and dropping
during subsequent cycles (Exhibit 22A, previous page). In real
terms, no particular regularity emerges (Exhibit 22B, previous
page), although this conclusion is subject to the caveat that
“real” (in this case) is measured using a backward-looking
measure of inflation; ex-ante calculations might differ slightly.
Given this backdrop, the famous bond market “conundrum” of
the 2000s does not look exceptional. Treasury yields during
that cycle followed a fairly typical sideways path in either
nominal or real terms.
As discussed earlier, short-term interest rates do tend to rise
as cycles proceed. As the Federal funds rate moves higher,
two-year bond yields get carried along. As a result, the
Treasury yield curve flattens during the course of the cycle.
This pattern has played out fairly consistently since the shift in
Federal Reserve behavior around 1980 and was particularly
evident in the 1990s and 2000s cycles, when Federal Reserve
rate hikes affected the short end of the curve much more than
the 10-year sector (Exhibit 23).
Today’s Treasury yield curve is steeper than usual for this stage
of the expansion
EXHIBIT 23: U.S. TREASURY CURVE SLOPE (10-YEAR/2-YEAR) BY CYCLE
350
Jul-80
300
Nov-82
250
Mar-91
Basis points
200
Nov-01
150
Jun-09
100
50
0
-50
-100
-150
-200
1
13
25
37
49
61
73
Months
85
97
109
121
Relative to previous cycles, the current expansion stands out
less for the behavior of the 10-year yield, which, while quite
low, has not behaved oddly in directional terms, and more for
the fact that short-term interest rates remain near zero five
years after the recession. As a result, the Treasury yield curve
is much steeper than usual for this stage of the expansion.
While we continue to expect longer-term yields (which currently
stand significantly below most estimates of U.S. potential
growth) to move higher in coming years, eventual curve flattening seems even more likely.
Conclusions
Five main conclusions emerge from the preceding examination
of business cycle history, theory and market relationships.
First, the age of the current expansion—five years in the U.S.,
less in the UK and euro area—in itself should not cause alarm.
Business cycles have been getting longer over time and, in any
case, do not appear to die merely of old age. Second, the current cycle does not seem particularly advanced, even in the
U.S. and certainly not in Europe. Most importantly on this
front, the level of interest-sensitive spending, especially on
consumer durables, remains quite low by long-term standards.
Moreover, credit growth has recovered only mildly from the
financial crisis. Third, financial markets seem likely to perform
reasonably well if the economy continues to avoid recession,
with equities likely to drift higher and credit spreads perhaps
tightening a bit further. Major, lasting reversals in these areas
outside of recessions have not occurred frequently. A tilt
toward growth-sensitive assets thus appears warranted, even
as the current expansion ages. The bond market has maintained a more ambiguous relationship with the business cycle,
although a flatter yield curve seems highly likely at some
point. Fourth, all of these views should be regarded as provisional. Business cycle dynamics and, in particular, recession
drivers remain poorly understood, and forecasters have struggled to predict even imminent recessions. Fifth, as a result,
institutional investors need to maintain robust hedge buckets
within their portfolios, even if they are optimistic about the
likelihood of a long-lived expansion. Assets that will perform
relatively well during a recession, such as government bonds
and high-quality corporate credit, represent an integral component of investing for the long run, given the difficulty in
predicting the timing of such downturns.
Source: J.P. Morgan; data as of June 2014.
J.P. Morgan Asset Management | 15 MARKET
INSIGHTS
PORTFOLIO
DISCUSSION:
Title Copy Here
Tracking
the
business cycle
AUTHOR
Michael Hood is Global Markets Strategist within the institutional
business at J.P. Morgan Asset Management (JPMAM). In this
capacity, he provides analysis of and commentary on the
economy and asset allocation to institutional investors of all
types. He writes frequent “Global View” commentaries, as well
as stand-alone publications on economic and market topics. He
also maintains forecasts for global growth, inflation and policy
interest rates, and contributes to the firm’s long-term capital
markets assumptions process.
Mr. Hood came to JPMAM in October 2011 from Traxis Partners, a
USD1bn+ macro hedge fund based in New York. There, he served
as chief economist from 2007 to 2011, maintaining detailed
forecasts for global variables. He produced a monthly global
outlook publication and frequent stand-alone pieces on a range
of developed and emerging market economic issues.
Previously, Mr. Hood worked as an economist and market
strategist at Barclays (within the emerging markets research
department) and at the J.P. Morgan investment bank (within
the economic research department). At J.P. Morgan, he began,
in 1994, as an economist for several Latin American countries.
Later, he oversaw J.P. Morgan’s Latin American economic
research effort and helped coordinate the department’s global
views. He contributed to and helped edit many J.P. Morgan
publications, including the weekly “Global Data Watch” and
quarterly “World Financial Markets.” He also created and edited
the quarterly “Latin American Economic Outlook” publication.
At Barclays, where he worked from 2004 to 2007, Mr. Hood
researched a combination of economic and market-strategy
topics within emerging markets, again writing for and helping
edit a variety of publications. While at J.P. Morgan and Barclays,
he frequently traveled to Latin America and spoke to a wide
range of clients, including institutional investors, corporations
and private equity sponsors.
Mr. Hood began his career in the research department at the
Federal Reserve Bank of New York, where he worked from 1992
to 1994 on a variety of international-finance and developingcountry topics. In this capacity, he wrote many country-risk
studies used by federal bank regulators.
NOT FOR RETAIL DISTRIBUTION: This communication has been prepared exclusively for Institutional/Wholesale Investors as well as Professional Clients as
defined by local laws and regulation.
The opinions, estimates, forecasts and statements of financial markets expressed are those held by J.P. Morgan Asset Management at the time of going to print
and are subject to change. Reliance upon information in this material is at the sole discretion of the recipient. Any research in this document has been obtained
and may have been acted upon by J.P. Morgan Asset Management for its own purpose. References to specific securities, asset classes and financial markets are for
illustrative purposes only and are not intended to be, and should not be interpreted as, advice or a recommendation relating to the buying or selling of investments.
Furthermore, this material does not contain sufficient information to support an investment decision, and the recipient should ensure that all relevant information
is obtained before making any investment. Forecasts contained herein are for illustrative purposes, may be based upon proprietary research and are developed
through analysis of historical public data.
J.P. Morgan Asset Management is the brand for the asset management business of JPMorgan Chase & Co. and its affiliates worldwide. This communication may
be issued by the following entities: in the United Kingdom by JPMorgan Asset Management (UK) Limited; in other EU jurisdictions by JPMorgan Asset Management
(Europe) S.à r.l.; in Switzerland by J.P. Morgan (Suisse) SA; in Hong Kong by JF Asset Management Limited, or JPMorgan Funds (Asia) Limited or JPMorgan Asset
Management Real Assets (Asia) Limited; in India by JPMorgan Asset Management India Private Limited; in Singapore by JPMorgan Asset Management (Singapore)
Limited; in Australia by JPMorgan Asset Management (Australia) Limited; in Brazil by Banco J.P. Morgan S.A.; in Canada by JPMorgan Asset Management (Canada)
Inc.; and in the United States by J.P. Morgan Investment Management Inc., JPMorgan Distribution Services Inc. and J.P. Morgan Institutional Investments, Inc.,
member FINRA/SIPC.
270 Park Avenue, New York, NY 10017
© 2014 JPMorgan Chase & Co. | MI_Tracking the business cycle
FOR INSTITUTIONAL/WHOLESALE OR PROFESSIONAL CLIENT USE ONLY | NOT FOR RETAIL DISTRIBUTION