The central
expectation is for the US recovery to continue to accelerate
throughout 2012, and we expect both the current consensus forecast of
2.2% in 2012 (up from 1.7% in 2011), and the US Federal Reserve's
band of 2.2% to 2.7% will prove to be excessively conservative.
The basis for that
judgement is that by our estimate returns on capital are around their
highest since 2000 and are still rising, which will continue to
foster investment spending; labour productivity continues to grow
(adjusted for changes in capital stock), which will underpin the
slowly- accelerating addition of jobs; and, most importantly, we
believe that the net develeraging of the economy which started in
2008 is now complete. We do not expect significant re-leveraging to
take place this year, but the mere fact that deleveraging is no
longer the key dynamic will shift the economy out of its modest 2.4%
annualized growth trend which it has sustained since the end of the
recession in 2009 and towards a 3%+ rate.
As this faster
growth path becomes acknowledged we expect to see bond yields lift
from their current excessively-low level (roughly 180bps below 'fair
value' in our models). It is the recognition of faster-than-expected
growth rather than a resurgence in inflation which we expect will
undermine the bond market.
This scenario has
threats both to the upside and the downside. On the upside, if
monetary velocity (GDP/M2) even stabilizes at its current precedent
low levels, then somehow we have to expect double digit nominal GDP
growth. On the downside, the cycle could be choked off by a
sustained rise in commodity prices sufficiently strong to erode the
US terms of trade sharply. What would it take? Oil at US$140 a barrel
would be threatening but not conclusive; oil at US$165 a barrel would
trigger a 'soft patch' to disrupt the recovery.
The fundamentals
supporting the business cycle remain unusually positive. Return on
capital is the most positive since 2000 and still rising, despite the
'soft patch' of mid-2011. Since aggregate capital stock is unlikely
to rise more than 1.5-2% this year in nominal terms, that return on
capital will keep rising. As a result, the capital spending cycle in
the US will continue to accelerate throughout 2012.
The allegedly 'jobless'
recovery will keep expanding payrolls at an accelerating rate. Real
labour productivity levels, adjusted for levels of capital per
worker, have recovered to levels last seen in 2001, and continue to
rise – albeit at a slightly slower pace. This real productivity
growth will continue to underpin growth in labour markets. In 2011,
employment rose by only 1% on average – in 2012 this can be
expected to rise to between 1.6% and 2.2%.
Most importantly, there
are signs that the household sector deleveraging which has been the
main driver of the US's sub-par recovery has burnt itself out. There
are both indirect and direct indications of this. The most indirect
are the way in which various debt to income, and net household
financial assets to GDP ratios have been returned to normal historic
levels.
The most direct are the
readings of monthly additions to consumer credit, which in recent
months have risen from previously steady readings of around US$7bn a
month to more than US$20-bn a month. But the most powerful
confirmation is the way in which the banking sector's loan to deposit
ratio first stabilized at around 81% in September (down from an early
2008 high of 102%), and have since very modestly begun once again to
expand.
We do not have to
assume that a structural deleveraging will immediately be followed by
a cyclical re-leveraging. But simply removing the deleveraging
dynamic will release one of the main breaks on the US cyclical
upswing. And this is consistent with what we see elsewhere in the
economy.
It is also consistent
with a continued fall in the US private sector savings surplus, which
fell to 3.8% in 2011 from 7% in 2010, and has generally been falling
at a rate of around 2.4percentage points a year. This will continue,
partly because bond yields have fallen so low as to represent
extremely poor value, with 10 year bonds yielding around 180 basis
points below what one would expect given policy rates, and prospects
for growth and inflation.
The Federal Reserve has
played a significant role in depressing bond yields, both directly
(through quantitative easing) and indirectly (through the
threat/promise of more quantitative easing to come, backed by the
release of extremely pessimistic growth and inflation forecasts).
Historically, one of the results of bond yields falling sharply below
'fair value' has been to dissuade saving; and conversely, when bonds
are 'cheap' relative to fair value, saving surpluses have tended to
rise.
From this we can reach
two conclusion.
- First, we should expect private sector savings surplus to continue to decline during 2012 – probably to around 1.5% of GDP.
- Second, the current over-valuation of US 10 year treasuries relative to 'fair value' make them extremely vulnerable to signs of the sort of accelerating US cyclical upturn we fully expect in 2012. What makes those bonds vulnerable is not necessarily 'the return of inflation', but rather a reassessment upwards of likely GDP growth, and reassessment downwards of the likelihood of third round of quantitative easing by the Federal Reserve.
Our central expectation
for the US, therefore, is for the cyclical upswing to gather momentum
throughout 2012. Moreover, this is now emerging in the monthly
data-runs, particularly in labour markets and surveys of business
conditions and consumer confidence, but also more cautiously in the
housing market and banking markets.
However, there remain
two main ways in which this central expectation can be blown off
course – one threat to the upside, and one to the downside.
Upside Risk:
Monetary Velocity and Nominal GDP
The threat to the
upside is simply stated: monetary velocity (GDP/M2) in 2011 sank to
lows unseen in the US since the end of the Second World War, yet at
the same time, M2 is growing around 10% a year with strongly positive
underlying sequential momentum. It therefore requires only that
monetary velocity falls no further for nominal GDP growth to start
accelerating into double-digit growth.
Two of the (linked)
factors which have depressed US monetary velocity are
- the deleveraging of the banking sector, as shown in the fall in loan/deposit ratios; and
- a sharp aversion by the household sector to financial risk since the onset of the financial crisis in 2007/08.
But we already can
observe that the fall in loan/deposit ratios is bottoming out. And we
also know that 2011's combination of extraordinary shocks
(principally from Japan and the Eurozone) is unlikely to have quite
the same power to surprise in 2012. In these circumstances, a
continuation of the sharp decline in monetary velocity is no
certainty.
Downside Risk:
Commodity Prices and Terms of Trade
The threat to the
downside is similarly easily stated: the US cycle remains vulnerable
to sharp and sustained rises in commodity prices. This vulnerability
shows up when we look at how both recessions and 'sort patches' in
recent US economic history have been preceded by a fall in the terms
of trade (caused by rising prices of imported commodities -mainly
oil).
Recessions
- 3Q00 to 3Q 01 – preceded by a 9% fall in terms of trade 1999- 3Q00.
- 1Q08 to 2Q09 – preceded by and intensified by a 13% collapse in terms of trade between 1Q07 to 2Q08.
Soft patches
- 2Q-3Q06 – preceded by a 7% fall in terms of trade between 1Q05 and 3Q05,
- 1Q-2Q11 – preceded by a 5% fall in terms of trade between 3Q10 and 2Q11.
Since the beginning of
4Q11 the rise in commodity prices has depressed US terms by nearly
3%, and the current rally in oil prices will extend that fall. As
that fall increases, so does the likelihood of a further 'soft patch'
emerging in the second half of 2012. In the absence of unexpected
severe supply-side shocks to rival the auto-industry shock delivered
by last March's Japanese earthquake/tsunami disaster, and given the
robust improvement in all other cyclical indicators, I would expect
the cyclical sensitivity to terms of trade falls to start to have an
effect nearer a 7% fall than last year's 5%.
The current spike in
oil prices (Brent at US$122 a barrel) won't do it. For reference, it
would take an oil price of US$165 a barrel to engineer a 7% fall from
the September terms of trade peak. Oil at US$140 a barrel would take
be consistent with a a 5% fall in terms of trade from the recent
peak.