Goldman Sachs: The global economy still has plenty of ‘room to grow’

Slack track: The global economy still has plenty of ‘room to grow’
Assessing size of output gaps allows us to anchor macro views
Global asset markets have re-priced substantially following hawkish policy
signals and perceived changes in the global economic outlook. Consensus
has built around the expectation that global growth will accelerate going into
2014, which should push output gaps in a positive direction. But there is less
clarity on where they currently stand or how they are likely to evolve. That is
key to forecasting policy paths and market reactions, as it allows us to
anchor macro views when markets fluctuate. So we look here at output gaps
in 40 countries, representing more than 90% of global output.
The global economy still has plenty of ‘room to grow’
We find that the global economy has plenty of ‘room to grow’, especially in
DMs, but also in some EMs. Our estimates point to gaps of around -2.4%
for the world, -3.4% for DMs and -1.0% for EMs. These estimates account
for possibly weaker potential growth rates after the crisis. During the next
2-3 years, output gaps are likely to close in EMs as cyclical weakness
recedes, and go half-way towards closing in DMs. But starting levels and
expected closing speeds differ markedly across countries.
Rising inflation as output gaps improve, but not everywhere
Inflation is likely to rise in places where output gaps improve the most,
with exceptions related to idiosyncratic factors. But unless inflationary
pressures become extreme, it is difficult to justify scenarios where policy
tightens financial conditions prematurely or excessively. When the recent
bout of market turbulence abates, markets should begin to reflect the view
that the world’s ‘room to grow’ is larger than is sometimes perceived, and
adapt to a gradual rather than abrupt return to normalcy.

�� -->

Slack track: The global economy still has plenty of ‘room to grow’
Assessing the size of output gaps—the difference between actual and potential GDP—is
key to forecasting the most likely path of macroeconomic policies, and the corresponding
market reactions. It is particularly helpful in times of heightened uncertainty, as it allows us
to anchor macro views as markets fluctuate.
In the past few weeks, global asset markets have re-priced substantially on the back of
seemingly hawkish policy signals. They have also responded to perceived changes in the
global economic outlook. Consensus has gradually built around the expectation that global
growth will accelerate in the second half of the year and into 2014. The engine behind that
pick-up is the US, as Jan Hatzius and team have described (see US Daily, ‘Moving Nicely
Over the Hump, But Not Yet Done’, May 21, 2013). But other developed markets (DM) are
also likely to grow faster, including Japan and the Euro area, as are some of the largest
emerging markets (EM).
Although that acceleration in growth should push output gaps in a positive direction, there
is less clarity with respect to where they currently stand, or how they are likely to evolve in
the near future. Because output gaps are not directly observable, not even after the fact
and not even to policy makers themselves, attempting to pinpoint them in a reliable way is
extremely useful. We examine here output gap measures for 40 economies, representing
more than 90% of global output. Our results can be summarised as follows:
 The global economy still has plenty of ‘room to grow’, especially in DMs, but also
in some EMs: Our estimates point to gaps of around -2.4% for the world, -3.4% for
DMs and -1.0% for EMs. These estimates account for possibly weaker potential
growth rates after the crisis, which would make output gaps less negative than
they would otherwise be. During the next 2-3 years, output gaps are likely to close
in EMs as cyclical weakness recedes, and go half-way towards closing in DMs.
 A more nuanced picture at the regional level: DM regions are submerged in very
negative output gap territory, especially North America (-4.1%), the Euro area
(-3.6%) and, to a lesser extent, Developed Asia/Pacific (-1.2%). EM regions have
smaller but also negative output gaps: Latin America (-0.7%), Emerging Asia
(-0.8%) and CEEMEA (-2.1%).
 Countries show relatively more dispersion in EMs: DM gaps are roughly split into
two tiers: some are deeply in negative territory (Euro area and US; at around 4.0%)
and some only mildly so (such as Japan and Canada, at around 1.0%). EMs are
smoothly distributed in a wider range: some (such as Mexico) have almost closed
gaps, some very negative gaps (Russia) and some are in between (China). Overall,
we expect most output gaps to turn less negative from now onwards, albeit at very
different speeds across countries.
 Rising inflation as output gaps improve, but not everywhere: As markets digest
recent policy signals, differences between countries will become more visible, and
these may be closely linked to output gaps. Inflation is one of the best examples of
where these differences can become most visible, as it is likely to rise as output
gaps improve—for example in Japan, Italy, the US and India—but fall owing to
idiosyncratic reasons in places such as Brazil, Russia and Spain. Ultimately,
markets should begin to reflect the view that the world’s ‘room to grow’ is larger
than is often perceived, and adapt to a gradual rather than abrupt return to
normalcy.

Estimating output gaps for the lion’s share of the global economy…
Output gaps (the difference between actual and potential GDP, expressed as a percentage
of the latter) are among the most relevant macroeconomic indicators. But they are also
among the most difficult to measure. This is because we do not observe potential GDP—
against which actual GDP is compared—and because there are varying definitions of what
any such measure should reflect. Broadly speaking, most economists think of it as the level
of output that would be consistent with full factor utilisation and the absence of inflationary
pressures. In that case, we say that the output gap is closed at zero. But most of the time
economies fluctuate around that level—going through periods of economic slack (with
negative output gaps) or overheating (with positive output gaps).
Output gaps are related to potential growth rates in an intuitive way: the only way to close
negative gaps is to increase actual growth rates above potential, at least for a while.
Otherwise, the economy does not have enough thrust to bring the level of GDP back to its
potential, and negative gaps will persist. Larger gaps require growth rates to surpass
potential by a significant margin and for a sustained period of time.
Determining the level of the output gap and how it is changing is a valuable exercise, as
this information may have a sizeable impact on policy actions, not only from the monetary
side but also from the fiscal side. The channel through which that occurs tends to be as a
result of concerns about missing inflation targets or even growth/unemployment targets,
which are directly affected by the degree of slack in the economy. The cross-country
aggregation of local output gaps can also shed light on the state of the global cycle, which
may have second-round effects on local slack conditions and asset returns.
We arrived at our output gap estimates by combining information from the IMF, the OECD
and our own models for 40 countries (half of which DMs and half of which EMs)
representing 94% of global GDP (see the box for more detail). Our sample starts in the
1980s for most economies, and we extended this through 2016 using our country
economists’ forecasts. But the focus of our analysis is the more recent period and the next
three years.
To compare countries and regions, our estimates follow a consistent approach. This has its
drawbacks, as our country economists often have specific information that allows them to
refine estimates of slack in particular ways. While those insights are valuable, there is also
value in attempting to use a consistent method across a broad range of countries. This may
not necessarily yield the best estimate for any given country, but it ensures comparability.
Because measuring output gaps is subject to uncertainty, it is also useful to combine
different approaches rather than select just one.
…Reveals somewhat lower potential growth rates after the crisis…
The first conclusion we draw from our output gap estimates is that the global financial
crisis had a differential impact on levels of slack across countries. The Euro area’s
sovereign crisis further accentuated those differences. Time will reveal the extent to which
those events had a permanent impact on potential growth rates—which may have come
from the dislocation of labour and capital markets or, more generally, the interruption in
the efficient functioning of the financial sector. In terms of labour productivity growth, we
have argued that the recent slowdown appears to be more related to the underutilisation of
inputs than to a permanent effect from the crisis (see Global Economics Weekly: 13/16,
‘Financial crises and productivity growth’, May 2013). But the implied potential growth
rates behind our output gap estimates do show a meaningful decline from before 2007 to
the period after 2010 in DMs (Exhibit 1) and to a lesser extent in some EMs

The growth rates compared in Exhibits 1 and 2 are measured over shorter windows than
those traditionally used to gauge potential growth, which tend to be about twice as long,
but they are an indication that a degree of structural slowing of attainable growth may
have characterised the post-crisis environment. The Euro area is the clearest example, with
potential growth rates estimated at roughly half of their pre-crisis levels. In contrast, the N-
11 and Latin America appear to show much more resilience. Overall, the important
observation is that the degree of slack implied by our output gap estimates across major
world regions does not arise from an overestimation of potential growth. If anything, the
evidence we find is that global slack persists even when accounting for possibly weaker
potential growth rates than before.
…And the persistence of substantial economic slack
Exhibit 3 shows our headline results in terms of the size and evolution of PPP-weighted
output gap estimates for the world, DMs and EMs. The building blocks are simply
differences in the level of GDP relative to potential GDP in each country, where that
potential follows the growth rates discussed in the previous section. Our main findings are
as follows:
 Negative output gaps persist around the world: After a bounce-back from the
global financial crisis, when economies moved towards closure of the gaps during
2010, gaps deteriorated again during 2011-12. At present, our estimates show
output gaps at around -2.4% for the global aggregate, -3.4% for DMs and -1.0% for
EMs. Of these aggregates, only EMs crossed into positive gap territory after the
crisis, while DMs have been navigating negative territory for five years in a row.
 There is relatively more ‘room to grow’ in DMs, but these new estimates suggest
there is probably slack in EMs too: We have argued for some time that the postbust
recovery of the global economy was likely to be sluggish, especially in DMs.
In relative terms, our results reflect that by showing a substantially more negative
output gap in DMs than in EMs. In turn, our results suggest that EMs—on average
at least—may no longer be as tightly constrained as we have sometimes worried
they are. This means that the period of below-potential growth since late 2010 may
have allowed some modest ‘room to grow’ there too, at least in some places.

 During the next 2-3 years, output gaps are likely to close in EMs and go half-way
there in DMs: Output gaps are likely to narrow. This should allow EMs to close
their gaps fully by mid-2016, and DMs to close half of their current gaps by late
2015. Mechanically, the only way to close output gaps is to see growth running
above potential rates for some time, and that is precisely what our forecasts for an
acceleration in growth starting later this year imply.
The picture is more nuanced at the regional level...
Our estimates allow us to distinguish the current state of output gaps across regions and
where they are likely to go in the medium run. This in turn helps to explain the different
macro trade-offs they face. This follows a simple framework we introduced last month,
where we assessed differentiation across countries based on those trade-offs, especially in
EMs as of late (see Global Economics Weekly: 13/19, ‘Macro trade-offs driving EM
differentiation’, May 2013). The basic idea is to compare key macro variables that matter
for policy and asset returns, one of which is the output gap.
DM regions are currently submerged in very negative output gap territory (Exhibit 4),
especially North America at -4.1% and the Euro area at -3.6%. Developed Asia/Pacific
(comprising Australia, Hong Kong, Japan, New Zealand, Singapore and South Korea) is
doing relatively better, at -1.2%. In fact, we expect it to be the first region to close its gap at
some point in late 2016. In contrast, the Euro area’s gap is likely to show substantial
sluggishness in the years to come, and North America is somewhere in between.
EM regions show much smaller but also negative output gaps (Exhibit 5), with Latin
America at -0.7% and Emerging Asia at -0.8%, and CEEMEA lagging behind at a deeper
-2.1%. In terms of their expected paths, the CEEMEA countries are likely to fall behind the
rest of EMs, while Emerging Asia moves relatively quickly to positive output gaps around
2015. Latin America is somewhere in between, with mildly negative output gaps that are
likely to close very slowly.

…As reflected by country-specific dynamics
At the country level, our estimates reveal an even more nuanced picture, especially in
terms of the level of output gaps and the speed at which they are likely to close. In the DM
group, Japan is not only closer to the zero line (-1.2%) but will also get there faster, almost
closing its gap by 2016 (Exhibit 6). In contrast, the negative output gaps in the Euro area (at
-4.1%) and the UK (at -3.8%) have no clear end in sight, while the US (at -4.4%) is likely to
show steady improvement over the next couple of years. In the EM group, there is a
dispersion of countries with almost closed gaps (e.g., Mexico, at -0.1%), those with
substantially negative gaps (e.g., Russia, at -2.4%) and countries in between (e.g., China, at
-0.6%) (Exhibit 7).
A positive implication from our estimates is that output gaps for the most part are likely to
improve from now through the end of 2014. Within the DM group, output gap levels are
currently split roughly in two tiers (Exhibit 8): some are deeply into negative territory (at
around -4.0%) and some are mildly there (at around -1.0%). The largest improvements in
the DM group are likely to be seen in Spain, the US, Japan and Italy, where output gaps are
expected to close by at least 80bp (Exhibit 9), and are currently among the lowest to begin
with. These dynamics accord with a general improvement in the investment climate that
we expect in some of these countries, especially in the US (as Noah Weisberger discussed
in Global Economics Weekly: 13/20, ‘Assessing the investment spending climate’, May
2013). But 11 more DMs should also see a positive change in their gaps, reflecting a
gradual tightening of capacity constraints, even if output gaps remain large.

Output gap levels in EMs are more smoothly distributed in a range from -6.0% to
somewhat above 1.0% (Exhibit 10). The largest improvements over the next year and a half
are likely to occur in Ukraine, the Czech Republic, India and Brazil, where we expect output
gaps to improve by at least 100bp (Exhibit 11). On the other hand, countries such as
Argentina, Chile, Thailand and the Philippines are likely to move in the opposite direction,
with downward movements in output gaps of at least 50bp. Our estimates already
incorporate recent downgrades of our growth views for various countries, more recently
for China, but the hit to growth from recent turbulence may extend to other EMs and pose
a downward risk to these expectations.

Rising inflation as output gaps improve, but not everywhere
From a market perspective, the main value in understanding where output gaps are at a
given point comes from their impact on policy actions and asset prices. Central banks that
find themselves in an environment of slack are more likely to ease than they would
otherwise—unless inflation pressures are too large. In turn, asset prices respond to those
policy decisions and to the economic dynamics associated with either positive or negative
output gaps. But policy actions and output gaps are somewhat endogenous, and that
endogeneity also affects asset prices, especially when market moves are as correlated as
they have been recently.
Because output gaps may be both cause and effect of a given policy stance, and asset
prices may price an expected change in output gaps well before it happens, it is difficult to
disentangle the links between them. Still, in the midst of recent market turbulence, output
gaps are one of the most helpful anchors we can use to pinpoint the state of the macro
picture across regions and countries. As markets digest the negative effects of the
withdrawal of Fed stimulus (see Dominic Wilson’s related discussion in Global Economics
Weekly: 13/21, ‘Higher US bond yields: Digestion, not congestion’, June 2013), fundamental
macro differences between countries will become more visible, and these may be closely
connected to output gaps.
The best example of that is the inflation outlook across countries. Exhibit 12 compares the
expected changes in output gaps with expected changes in yoy inflation (excluding
Argentina, Ukraine and Venezuela, where inflation dynamics are atypical). In most
countries where output gaps are likely to improve, inflation is also likely to rise. This
applies to some of the cases we identified as having the largest expected improvements:
Italy, US, Czech Republic and India. The exceptions are Russia and Spain, which should see
improvements in output gaps but also a reduction in inflation, from currently high levels in
the former and owing to deflationary forces in the latter. As we have argued elsewhere in
discussions related to macro trade-offs, what matters ultimately in terms of inflation is the
deviation with respect to ‘target’. But as the economic outlook normalises, and DMs exit
liquidity traps, the usual connections between output gaps and inflation are likely to
strengthen in the direction the simple scatter implies

Output gaps are therefore fundamental pieces of our macro outlook. And they suggest that
the global economy still has plenty of ‘room to grow’, especially in DMs, but also in some
EMs. With that in mind, unless inflationary pressures become extreme—an unlikely
outcome in most DMs given the degree of slack—it is difficult to justify scenarios where
policy tightens financial conditions prematurely or excessively. Some EMs may find they
face more complicated macro trade-offs, and this could explain some of their recent
underperformance (see Kamakshya Trivedi and George Cole’s Emerging Markets Macro
Daily, ‘Assessing the EM sell-off through a financial conditions lens’, June 11, 2013). Of
course, although far from our baseline outlook, there is always the possibility that a crisis
scenario unravels, especially in those EMs where pressure may become too intense. That
said, when the recent bout of market turbulence abates, markets should begin to reflect the
view that the world’s ‘room to grow’ is larger than is sometimes perceived, and adapt to a
gradual rather than abrupt return to normalcy.
José Ursúa