Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.
The following is in response to Robert Huebscher’s article, The New Challenges to Reinhart and Rogoff, which appeared last week:
Your article contained the following table:
Average real GDP growth rates by debt-to-GDP category
|
Under 30%
|
30%-60%
|
60%-90%
|
Over 90%
|
|
Reinhart and Rogoff
|
4.1
|
2.9
|
3.4
|
-0.1
|
|
U. Mass.
|
4.2
|
3.1
|
3.2
|
2.2
|
It revealed the following historical relationships:
-
High (above average) real GDP growth rates have been associated with countries with low to modestly (<30% debt-to-GDP ratios) leveraged governments,
-
Moderated (average) real GDP growth rates have been associated with countries with moderately to highly (30% to 90%) leveraged governments, and
-
Subdued (below average) real GDP growth rates have been associated with countries with extremely (> 90%) leveraged governments.
Interestingly, the three observations listed above hold true in all three observable cases, i.e., with a) Reinhart and Rogoff (R&R) data only, b) with UMass data only, and c) any blend of R&R and UMass data.
This is not overly surprising data and likely not wholly inconsistent with R&Rs primary meta-message.
These readily observable associations, as you point out, do not necessarily lead to a sound conclusion of causality, much less indicate the direction of any possible causality.
All the academic data torturing in the world can, and I'm sure ultimately will be done on this near infinitely broad, complex, and highly dynamic subject. And, I suspect that due to the highly complex and always moving/morphing nature of the target topic, virtually every seriously studied conclusion reached will remain of the transient and ultimately refutable variety. Holy Grail-like models can be difficult to attain and especially to retain, period… not just in the realm of global macroeconomics.
Absent clear, irrefutable and broadly accepted and therefore confident policy-implementable ”Graildom” around this truly huge question of "Precisely how does degree of and/or change of degree of government leverage impact a country's real GDP growth rate variability, and with what, if any, leads and lags?," we can likely benefit from harking back to foundational principles. You started to do so with your reference to the finance axiom that viable projects require expected returns in excess of their expected costs (including financing costs).
In that vein, I posit further, and even more fundamentally, that in a sovereign like ours, with very thoughtfully derived and explicitly imposed constitutional limits (20th and 21st century gross infringements thereon aside) on government, and with express emphasis on protection of private property rights, baseline filtering on such projects could be most constructively undertaken in the following manner and order of priority:
-
First, discern legality. Apply the constitutionality/legality filter, the one that Speaker Pelosi relatively recently and so (in)famously expressed disgusted disbelief over. Her very reaction should make the point of the degree of the disdain for our (at least theoretically) rule-of-law based system. We should first, foremost and always seriously ask the basic question of "Does this resource-requiring undertaking (the term ’investment’ seems, for rhetorical purposes I suppose, to be used quite loosely here) even fall within the purview of the federal government to begin with? In other words, is it legal? Not rationalizably legal, under conjured/tortured interpretations (or hoped for future interpretations) of the law, but is it clearly and obviously heads-up legal on its face pursuant to our U.S. Constitution, as amended to date?
Next, if it actually passes a legitimate legality test,
-
Clarify whose proper purview it optimally belongs under economically – should it be undertaken by the public or private sector? The testing for this is properly done under the area of economics called publicfinance, whereby legitimately public goodsand services are analytically ascertained and distinguished from private sector goods and service. By definition, in order to be classified in economic terms as a public good or service, a good or service must be both non-excludable (everyone can access/benefit) and non-rivalrous (access/benefit is not zero-sum).
Assuming more than one proposed project legitimately gets through both of the above screens, then, because government resources are, even in a world where borrowing is possible, less than infinite, prioritization/triaging is indicated.
- This final cut step is where the Huebscherian notion of applying traditional cost/benefit financial break-even analysis logically comes into play. The difficulty of course in applying this financial analysis is the complexity of the real world where we have to bring into the analyses, for example:
- Estimates of the costs of externalities
- Assessments of the costs/benefits of unintended consequences
- Displacement/opportunity costs
Beyond the sheer complexity of the analysis, the really hard fact to deal with is that costs/returns are always and forever in this context going to be estimated/forecasted cost/return numbers, not actual numbers. Further, when these cost versus return analyses are made under the heat of the political context and pressures associated with prospective government expenditures, the underlying assumptions will logically be biased and distorted by the environment. And their very scale tends to incentivize and thus ramp up the amount of such assumptions distortion. The biases and distortions are simply endemic.
While the idea of finally and definitively nailing down some perfectly or near perfectly deployable model of multiplier analysis, etc. to causally connect a government's budgetary urges and antics with its country's economic growth prospects is hugely appealing (even to me) from an intellectual standpoint, and likely sends an exciting tingle up many a ruling-class leg, it should not be thought of by serious people as a Holy Grail policy tool for government spending policy purposes. It is a mere (possibly discernible) tree in a forest -– a forest the three primary and dominant giant redwoods of which are the three filters described above. Is it legal? Who should properly be doing it? And, what's most viable?
Those three things, and in that order, comprise the core essence of the forest I'm concerned gets too readily obscured by the siren song of a theoretically really nifty econometric modeling tree that, if we can just finally wrestle it to ground, will cough up magical policy prescriptives and of course, economic Nirvana!
I loved your article!
As always, thanks for doing what you do, and especially for doing it so well.
Guy Cumbie is the principal ofCumbie Advisory Services, a Fort Worth, TX-based advisory firm.
Read more articles by Guy Cumbie
