Last week's GDP report certainly allayed conventional wisdom's fears that the economy was sliding back into recession. The headline number for personal consumption expenditures (PCE) was the clincher since it supposedly showed that the consumer, for all the worry about prices, is back and supporting the consumption economy. Sixty-nine percent of third quarter GDP growth came from additional spending on PCE; 80% of that PCE was due to increases in consumer spending on "services".
For now we will ignore the alarmingly large 1.7% decline in real disposable personal income and the obvious slowing of GDP when viewing the numbers year-over-year (stripping out seasonal adjustments) to simply focus on how these headline numbers might fit into the basic ideas of a recovery and a healthy economic system.
Everyone wants to believe that the recovery is real and unfaltering since the alternative is, to far too many people, especially policymakers, unfathomable. But the stakes here are too high to ignore compositions. The most robust segments of spending on personal services were health care and housing. Those two alone accounted for 39% of all GDP growth in Q3, and 57% of the "robust" consumer segment.
The measurement of consumer spending on housing services is far more controversial than most people realize, and it gets right to the idea of what we should be trying to measure by all these numbers. The largest component of PCE spending on housing services (and the largest single piece of overall GDP) is something called the "imputed rental value of owner-occupied housing". The Bureau of Economic Analysis (BEA) estimates how much homeowners who live in their own homes would pay themselves in rent during a certain time period.
Think about that for a moment. Nearly $1.2 trillion of total GDP comes from an estimate of how much a homeowner would pay himself in rent should that homeowner ever want to do such a ludicrous thing.
It turns out, however, that this phantom GDP component has pretty sound technical reasoning. The BEA is trying to ensure that GDP is "invariant to how certain activities are carried out". In other words, if the BEA did not include the phantom imputed rental value of owner-occupied housing, GDP, as it is currently calculated, would decline during periods of increasing home ownership. Since only rental income flows into the GDP accounts, more people buying and owning houses likely means less real rental activity, leading to a statistical decline in the GDP calculation.
The keepers of the economic accounts believe that home ownership is largely a benefit to society, so it makes some sense to have the GDP calculation reflect a beneficial trend and become invariant. The process of compiling economic data is meant to show that the measurement of goods and services produced is an accurate reflection of an improving or worsening economy. Reflecting a growing beneficial trend by conjuring a phantom number fits within this overall schematic.
But just as home ownership is thought to be universally beneficial, the housing bubble itself demonstrated an opposing case. There can be no argument that a large proportion of economic activity predicated on this socially positive trend ended up being a nightmare. But the economic accounts make no qualitative distinction between any kind of activity, including the housing imputation. All economic activity is believed, imputed or real, to be an accurate measure of a positive economic trend, so home ownership is accepted as an economic positive without further question or qualification.
That leads to an important distinction about what an economy really is. Is it really just about the quantity of goods and services produced and exchanged? If we regard the imputation of owner-occupied housing rent as valid, then what is actually being accomplished economically by owning (not creating) houses? This question is somewhat captured in the technical definition the BEA uses to explain its rationale: it is trying to measure the value of the service of "shelter".
There is clearly some kind of "good" in the concept of shelter, but I wonder if it is really a true economic concept. I counter the economic value of shelter by using an example of someone that builds their own home from scratch, right down to creating and fabricating all the material (think of a rustic cabin in the woods). That person has obviously created a means to access the social good of providing shelter, but has done so only for himself. There is no economic exchange in the procedure and therefore no real benefit to anyone in the larger economic system. But since that made-from-scratch house was actually made, it will be included in the GDP report once the BEA includes that person within the owner-occupied rent survey. Similarly, there is no exchange of the shelter "service" when one buys a house to live in, even though the "service" of shelter is being accessed or used.
The creation of goods or phantom services cannot be the sole object or arbiter of economic success. Rather, it is the process of exchange, particularly the pathway of exchange, that creates the economic "good". In exchanging goods and services, an economy that allows for labor specialization, leading ultimately to increasing productivity, defines economic success. If we all had to make our own houses and grow our own food to eat we would have little time to do anything else. The modern lifestyle, itself a social "good", is based on nothing more than expanding productivity through labor specialization.
If I build my own home and grow my own food, the economic statistics should necessarily reflect degradation in the wider economic system. If I do more for myself, then specialization is reversing and breaking down, meaning someone somewhere will not be able to specialize (to the same degree) as I withdraw from the system of exchange. But the overall quantity of goods produced is relatively similar. Should it not follow, then, that a positive or growing economy should be properly defined as a system of broadening labor specialization, perhaps independent of the quantity of goods produced? In terms of the service of shelter, unless I purchase it from someone it does not really have an economic component, even though it does exist.
Manufacturing jobs have moved overseas over the last ten years as the dollar has been devalued. That was not a pressing problem since the service sector, especially health care, education and "business services", filled in some of the gap. The majority of that gap was bridged by the "wealth effect" of rising asset prices leading to debt accumulation that fueled consumer spending (showing up as the near-zero savings rate). It follows from that that the beneficial cycle of labor specialization was increasingly based on asset prices and debt accumulation expressed through construction and service jobs, but also through increasingly imported goods where trade dollars flowed back to the U.S. as GSE debt (a unilateral system of trade where labor specialization was indirectly caused by the methodology of money circulating to households through real estate prices and debt, a cycle that no longer exists).
Jeffrey Snider is President and Chief Investment Officer of Atlantic Capital Management, a registered investment advisor.
Read Full Article »