One of today’s great paradoxes is the perceived lack of spillovers. Macro theory stresses interrelationships within economies, between markets, and across borders. Yet most financial market participants now believe in the theory of containment – that disruptions in one sector, one market, or even one economy can, in effect, be walled off from the rest of the system. Whether it’s the bursting of the US housing bubble, carnage in sub-prime mortgage lending, or a slowing of Chinese investment, these events are quickly labelled as “idiosyncratic” – unique one-off disturbances that are perceived to pose little or no threat to the larger whole. The longer a seemingly resilient world withstands such blows, the deeper the conviction that spillover risk has all but been banished from the scene. Therein lie the perils of a dangerous complacency.
Spillover from the US housing bubble
The post-housing-bubble shakeout of the US economy is an important case in point. There’s little disagreement on the wrenching adjustments that have already unfolded in this sector – a 25% drop in new home sales, a 35% plunge in housing starts, a 16% annualized decline in homebuilding activity over the past three quarters, and a reduction of 110,000 jobs in the residential construction industry from its recent peak. But the broader US economy barely flinched – at least, so far. Even in the face of a mild slowdown of 2.3% annualized real GDP growth in the two middle quarters of 2006, the economy still expanded by 3.4% over the four quarters as a whole. Such are the footprints of what many call the “two tier” economy – a weak housing sector (maybe autos too) accompanied by persistent resilience elsewhere.
I don’t question the facts as they have unfolded – other than noting the obvious distortions to seasonally adjusted construction data during periods of unusually warm winter weather. What I do question is the conclusion that this shakeout is not likely to have significant spillover effects on the broader US economy. The most obvious candidate remains the seemingly invincible American consumer, whose real spending growth accelerated to a brisk 4.4% annualized clip in the final period of 2006. I continue to cling to the seemingly discredited notion that it’s only a matter of time before the consumer responds to the carnage in the US housing market. The bulk of the income effects are yet to come – especially since the employment declines in residential construction have unwound only about 14% of the hiring boom of over 810,000 that occurred in this sector over the five years ending January 2006. As building activity is brought down into alignment with now depressed rates of new home sales and housing starts, commensurate adjustments can be expected in labour input – and in the income generation driven by such employment. If that occurs, consumers will be squeezed by the coming housing-related shortfall of purchasing power.
Pressure on American consumers
Asset effects are also likely to keep putting pressure on American consumers. A personal saving rate that has now been in negative territory for two years in a row leaves little doubt of the asset-dependent support to US consumption. With nationwide house price appreciation now slowing dramatically – from 13.9% y-o-y in 2Q04 to 7.7% in 3Q06 on an OFHEO basis – and likely to slow a good deal further in the months ahead, consumers will have considerably less in the way of excess asset appreciation that can be used to support spending and saving. Net equity extraction from residential property has already fallen from 8.5% of disposable personal income in late 2005 to 6.5% in late 2006. That’s especially worrisome with debt ratios at record highs and income-based saving rates at record lows.
Looking through the noise of energy-related gyrations to headline prices and inflation-adjusted household purchasing power, I remain highly sceptical of the consumer resilience call in a post-housing-bubble climate. And if the consumer finally fades, as I suspect, capital spending will be quick to go as well. I draw no comfort from ever-abundant coffers of corporate cash flow. If the demand outlook turns shaky due to spillover effects from housing to consumption, businesses will rethink expectations of future pressures on capacity utilization – and cut back plans to expand capacity accordingly. The recent softening of capital goods orders – average declines of -0.9% in nondefense capital good bookings ex aircraft in the final three months of 2006 – is especially supportive of that conclusion.
There is also an important financial dimension to the spillover debate – underscored by the rapidly evolving carnage in America’s sub-prime mortgage lending business. Like virtually every other credit event that has unfolded in the past several years – from auto downgrades to the implosion of Amaranth – our credit strategists have been quick to label the sub-prime mortgage problem as idiosyncratic. While spreads have blown out in this relatively small segment of the US mortgage market – with sub-prime loans about 11% of total securitized home loans – spreads for higher rated mortgage credits have been largely unaffected.
Again, I don’t dispute the facts as they have unfolded so far. My problem comes in extrapolating this resilience into the future. With resets on floating rate mortgages likely to put debt service obligations on a rising path for already overly-indebted US homeowners, the case for increased default rates and collateral damage on prime mortgage lenders looks increasingly worrisome. Indeed, as the recent warning from HSBC just indicated, it’s not just the small specialized lenders that are now being hit. Spillover effects are quickly moving up the quality scale on the financial side of the post-housing-bubble shakeout story, and their potential for impacts on the broader economy can hardly be dismissed out of hand.
The risk of a Chinese investment slowdown
Halfway around the world, a comparable issue is evident with respect to the Chinese investment slowdown. Like America’s housing shakeout, there can be little disputing the facts of a major slowing of Chinese investment activity – a year-over-year growth rate that was running at close to 30% at the start of 2006 but that ended the year at 14%. Despite this dramatic slowing, most still believe nothing can stop China’s growth juggernaut. However, with investment easily the largest sector of the Chinese economy – close to 45% of total GDP in 2006 – it is almost mathematically impossible for sharply slower investment growth not to have impacts on the broader economy. The recent industrial output trajectory underscores this conclusion – a slowing from peak rates of growth of 19.5% last June to less than 15% in the final months of 2006. While 15% growth in industrial output is still quite vigorous, it does represent a meaningful cooling off from earlier overheated gains.
At the same time, I take the recent softening of commodity markets as further validation of the spillover effects of China’s investment slowdown. With China accounting for about 50% of the cumulative increase in global consumption of base metals and oil since 2002 – fully 10 times its 5% share of world GDP – a China slowdown represents a very important development on the demand side of economically sensitive commodity markets. The same can be said for the transmission of spillover effects into China’s supply chain. As Chinese investment slows, cross-border impacts are likely in the other big economies of Asia – especially Japan, Korea, and Taiwan. Similar ripple effects should be felt by China’s natural resource providers – especially Australia, Brazil, Canada, and parts of Africa. In recent years, China has become such an important engine on the supply side of the global economy that it is difficult to see how a meaningful deceleration in its major source of economic growth won’t produce significant collateral damage elsewhere in the world.
Modern-day macro is a theory of interdependence – linkages both within and between economies. As such, spillovers are the norm, not the exception. Consequently, if an economy is hit with a major shock – like the bursting of the US housing bubble or a cooling off of China’s investment surge – it is very difficult to contain the damage before it spreads elsewhere in the global macro system. The best containment strategy is autonomous support to internal demand – especially private consumption. But even in those cases, it would take an acceleration of growth in the resilient sector(s) to offset the impacts of slower growth in the shocked sector. In my view, that’s highly unlikely for either the US or China.
Geopolitical spillover: global flashpoints
The biggest risk of all may be the geopolitical spillover. At a recent Morgan Stanley client conference, Middle East security expert Kenneth Pollock underscored the risk of cross-border spillovers in the aftermath of civil wars. He argues that was true in the case of recent civil wars in Afghanistan, the Congo, Lebanon, Somalia, and Yugoslavia, and could well occur in response to the current civil war in Iraq (see Daniel Byman and Kenneth Pollack, “Things Fall Apart: Containing the Spillover From an Iraqi Civil War,” Brookings Institution Analysis Paper Number 11, January 2007). Pollack warned that as Iraq veers out of control, pan-regional spillover effects in Iran, Israel, Saudi Arabia, Jordan, Kuwait, and Turkey could well be unavoidable. Such a possibility could not only further destabilize an already volatile part of the world but could wreak havoc with oil prices and the broader global economy.
Financial markets have learned to shrug off spillover risks in recent years. An ample cushion of excess liquidity has been key in defusing the potential impacts of massive current account imbalances, soaring oil prices, the bursting of the equity bubble, and an escalation of terrorist activity. Most investors are now of the view that spillover risk is inconsequential for such a Teflon-like world. History does not treat such complacency kindly
By Stephen Roach, global economist at Morgan Stanley, as first published on Morgan Stanley’s Global Economic Forum