A few days ago, Scott Sumner blogged about the “new normal” of NGDP trend (3% a year) (here and here). Overall, I tend to agree with him that the aggregate nominal expenditures are now at a new and historically low trend growth rate. I think that he is way too optimistic!
As readers of this blog are aware, I am not convinced that NGDP is the proper proxy for nominal expenditures. I believe that Nominal Gross Output (NGO) is a better proxy as it captures more goods and services traded at the intermediate level (see blog posts here, here and here). The core of my argument is that NGO will capture many “time to build” problems that will not appear in GDP as well as capture intangible investments which are now classified otherwise (see literature on intangible investment as capital goods here). Thus, my claim that NGO captures more expenditures (especially between businesses). (Note: I am in the process with some colleagues of finalizing a paper on the superior case for NGO – more on this later.)
Are we at a new normal point where growth in nominal expenditures is slower than in the past? Yes! But according to Sumner and others, this is 3%. If we use NGO we are much lower! Again, using the FRED dataset, here is the evolution of NGO and NGDP since January 2005 (the start date of the NGO series in a quarterly form). The graph shows something odd starting in mid-2014 : NGO is growing much more slowly.
To see this better, let’s plot the evolution of NGO as a percentage of NGDP in the graph below. If the ratio remains stable, then the trend is similar for both. As one can see, the recession saw a much more pronounced fall of NGO than NGDP with a failure to return to the initial levels. And since 2014, the ratio has started to fall again (indicating slower growth of NGO than NGDP).
So what is happening? Why is there such a difference? I am not one hundred percent sure about the causes of this difference. However, I am willing to contend that NGO is fitting better than NGDP with other indicators indicating a tepid recovery. If we look at the labor force participation rate in the US, it continues to fall in a nearly mechanical manner. Fewer and fewer workers are at work (or looking for work) in comparison to the population that could be working while investments are disappointing.
Maybe Scott Sumner is being overly optimistic. The new trend might simply be substantially lower than he believes.*
*Readers should note that I believe that monetary policy is, at present, too restrictive. However, I believe the culprit is not the Federal Reserve but financial regulations that restrict the circulation of “private money” (the other components of broad money found in divisia indices – see my blog post here).
A few weeks ago, I finished reading Scott Sumner’s The Midas Paradox. As an economic historian, I must say that this is by far the best book on the Great Depression since the Monetary History of the United States. Moreover, it is the first book that I’ve read that argues simply that the Great Depression was the result of a sea of poor (and sometimes good) policy decisions. However, coming out of the book, there was one thing that came to mind: Sumner is underselling his (very strong) case.
In essence, the argument of Sumner looks considerably like that of Milton Friedman and Anna Schwartz: The Federal Reserve allowed the money supply to contract dramatically up to 1932, turning what would have been a mild recession into a depression. However, Sumner adds a twist to this. He mentions that after the depth of the monetary contraction had been reached, there was a reflation allowing an important recovery during 1933. This is standard AS-AD macro of a (very late) expansionary policy to allow demand to return to equilibrium. Normally, that would have been sufficient to allow the rebound. Basically, this is the best case for NGDP targeting: never let nominal expenditures fall below a certain path because of a fall in demand. The problem, according to Sumner, is that the recovery was thwarted by poor supply-side policies (like the National Industrial Recovery Act, the Agricultural Adjustment Act etc.). The positive effects of the policy were overshadowed by poor policy. And thus, the depression continued.
To be fair, Sumner is not the first to emphasize the “real” variables side of the Great Depression. I am especially fond of the work of Richard Vedder and Lowell Galloway, Out of Work, which is a very strong candidate for being the first econometric assessment of the effects of poor supply-side policies during the Great Depression. I was also disappointed (but not too much since Sumner did not need to make this case) to see that no mention was made of the Smoot-Hawley tariff as a channel for monetary transmission (as Allan Meltzer argued back in 1976) of the contraction. Nonetheless, Sumner is the first to bring this case so cogently as a story of the Great Depression. Thus, these small issues do not affect the overall potency of his argument.
The problem, as I mentioned earlier, is that Sumner is underselling his case! I base this belief on the experience of England at the same time. Unlike the United States, the British decided to apply their piss-poor supply-side policies during the 1920s – well before the depression. The seminal paper (see this one too) on this is by Stephen Broadberry (note: I am very biased in favor of Broadberry given that he is my doctoral supervisor) who argued that the supply shocks of the 1920s caused substantial drops in hours worked and although the rise of unemployment benefits played a minor role, the vast majority of the causes were due to the legal encouragement of cartel formation. As a result, there were no supply-side shocks during the depression to create noise. However, England did have a demand-side expansionary policy in 1931. Even if it was by accident more than by design, England left the gold standard in September 1931. This led to the equivalent of an easy monetary policy and the British economy stopped digging and expanded afterwards. The Great Depression was not a pleasant experience for the British, but it was not even close to the dreadful situation in the United States. As a result, we can see whether or not it was possible to exit the Great Depression by virtue of a monetary policy. I’ve combined the FRED dataset on monthly industrial production and the monthly GDP estimates for inter-war Britain produced by Mitchell, Solomou and Weale (see here) to see what happened in England after it left the gold standard. As one can see, the economy of Britain rebounded much more magnificently than that of the United States in spite of supply-side constraints.
Sumner should expand on this point! To be fair, he does talk about it briefly. Not enough! A longer discussion of the British case provides him with the “extra mile” to cover the distance against competing theories. The absence of supply shocks in Britain during the Depression confirm his story that the woes of the United States during the 1930s are due to initially poor monetary policy and then poor supply-side policies. In my eyes, this is a strong confirmation of the importance of the NGDP level target argument!
With such a point made, it is easy to imagine a reasonable counterfactual scenario of what economic growth would have been after monetary easing in 1933 in the absence of supply-side shocks. Had the United States kept very unregulated labor and product markets, it is quite reasonable to believe (given the surge seen in 1933 in the Industrial Production data) that the United States would have returned to 1929 levels. In the absence of such a prolonged economic crisis, it is hard to imagine how different the 1930s and 1940s would have been but it is hard to argue that things would have been worse.
UPDATE: From the blog Historinhas, Marcus Nunes sent me the graph below confirming the importance of the NIRA shock on eliminating all the benefit from easy money after 1933.
A week ago, I initiated a discussion on using another indicator of nominal spending instead of NGDP when the time comes to set monetary policy. My claim was that NGDP includes only final goods and as a result, it misses numerous business-to-business transactions. This means that NGDP would not be the best indicator. I propose a shift to a measure that would capture some intermediate transactions.
The result was a response by Nick Rowe (to which I did respond), Matt Rognlie, Marcus Nunes and Scott Sumner (to whom I am responding now). Nunes and Sumner are particularly skeptical of my claim. I am providing a first response here (and I am attempting to expand it for a working paper).
The case against NGDP
GDP has important shortcomings. First of all, thanks to the work of Prescott and McGrattan (2012 : 115-154), we know that a sizable part of capital goods acquisition fails to be included inside GDP. That sizable part is “intangible capital” which Prescott and McGrattan define as the “accumulated know-how from investing in research and development, brands, and organizations which is the most part expensed rather than capitalized” (p.116). Yet, investments in research and development are – in pure theoretical terms – like the acquisition of capital goods. However, national accounts exclude those. Once they’re included in papers like those of Prescott and McGrattan and those of Corrado, Hulten and Sichel (2009), increases in productivity were faster prior to 2008 and that the collapse after 2008 was much more pronounced. In addition, this form of capital is increasing much faster than tangible so that its share of the total capital stock increases. Thus, the error of not capturing this form of capital good investment is actually growing over time causing us to miss both the level and the trend.
A second shortcoming of importance is the role of time in production. Now, just the utterance of these words makes me sound like an Austrian. Yet, this point is very neoclassical since it relies on the time to build approach. In the time-to-build model of the real business cycle approach, production occurs over many periods. Thus changes in monetary policy may have some persistence. The time-to-build model proposes that firms undertake long projects and consume more inputs. In terms of overall transactions, this will mean more and more business to business (B2B) transactions. Hence if an easy monetary policy is inciting individuals to expand their number of projects that have more distant maturities, then a focus on GDP won’t capture the distortionary effects of that policy through. Similarly, if monetary policy tightens (either directly as a fall of the money supply or through an uncompensated change in velocity), the drop in economic activity as projects are closed down will not equally well captured. While this point was initially advanced by Kyland and Prescott (1982), some Austrians economists have taken up the issue (Montgomery 1995a; 1995b; 2006; Wainhouse 1984; Mulligan 2010), several neoclassicals have also taken it up (Kühn 2007; Kalouptsidi 2014; Kyland, Rupert, Sustek, 2014).
Why shift to another measure
My contention is that NGO (Nominal Gross Output) allows us to solve a part of that problem. First of all, NGO is more likely to capture a large share of the intangible capital part since, as a statistic, it does not concern itself with double counting. Hence, most of the intangible capital expenses are captured. Secondly, it also captures the time-to-build problem by virtue of capturing inputs being reallocated to the production of projects with longer maturities.
Thus, NGO is a better option because it it tries to capture the structure of production. The intangible capital problem and the time to build problem are both problems of intermediate goods. By capturing those, we get a better approximate idea of the demand for money.
Let me argue my case based on the Yeager-esque assumption that any monetary disequilibrium is a discrepancy between actual and desired money holdings at a given price level. Let me also state the importance of the Cantillon effects whereby the point of entry of money is important.
If an injection of money is made through a given sector that leads him to expand his output, the reliability of NGDP will be best if the entry-point predominantly affects final goods industry. If it enters through a sector which desires to spend more on intangible investments or undertake long-term projects, then the effects of that change will not appear as they will merely go unmeasured. They will nonetheless exist. Eventually firms will realize that they took credit for these projects for which the increased output did not meet any demand. The result is that they have to contract their output by a sizable margin. In that case, they will abandon those activities (imagine unfinished skyscrapers or jettisoned research projects).
In such situations, GO (or even a wider measure of gross domestic expenditures) are superior to GDP. And in cases where the effects would start in final-goods industry, then they have the same efficiency as GO (or the wider measure of gross domestic expenditures.
The empirical case
The recurring criticism in most posts is that NGO is volatile over the period when the data is available (2005Q1-today). True, the average growth rate of NGO is the same as NGDP over the same period, but the standard deviation is nearly twice that of NGDP. However if you exclude the initial shock of the recession, the standard deviations converge. In a way, all the difference in volatility between the two series is driven by the shock of the recession. Another way to see it is to recompute two graphs. One is an imitation of the graphs by Nunes where NGDP growth in period T is compared with growth in the period T minus 1, but we add NGO. The second is the ratio of NGO to NGDP.
As one can see from the first figure, NGO and NGDP show the same relation except for a cluster of points at the bottom for NGO. All of those lower points are related to the drop from the initial recession. All concentrated at the bottom. This suggests that the recession had a much deeper effect than otherwise believed. The second graph allows us to see it.
The ratio of NGO to NGDP shows that the two evolved roughly the same way over the period before the recession. However, when the recession hit, the drop was more important and the ratio never recovered! This suggest a much deeper deviation from the long-term trend of nominal spending which is not seen at the final level but would be seen rather in the undertaking of long-term projects and the formation of intangible capital (the areas that NGDP cannot easily capture).
The case for NGO over NGDP is solid. It does not alter the validity of the case for nominal spending stability. However since the case for nominal spending stability hinges on total transactions of inputs and outputs more than it does on the final goods sold, NGO is a better option.
Quick comment in response to Rognlie
In his reply to Nick Rowe, Matt Rognlie states that the more important fall of NGO is explained by changes in relative prices. Although his transformation shows this, the BEA disagrees. Here is the explanation provided by the BEA:
For example, value added for durable-goods manufacturing dropped 15 percent in 2009, while gross output dropped 19 percent. The decline in gross output is much more pronounced than the decline in value added because it includes each of the successive declines in the intermediate inputs supply chain required to manufacture the durable goods.
I did not think that my post on NGO versus NGDP would gather attention, but it did (so, I am happy). Nick Rowe of Carleton University and the (always relevant) blog Worthwhile Canadian Initiative responded to my post with the following post (I was very happy to see a comment by Matt Rognlie in there).
Like Mr. Rowe, I prefer to speak about trade cycles as well. I do not know how the shift from “transactions” to “output” occurred, but I do know that as semantic as some may see it, it is crucial. While a transaction is about selling a unit of output, the way we measure output does not mean that we focus on all transactions. I became aware of this when reading Leland Yeager (just after reading about the adventures on Lucas’ Islands). However, Nick (if I may use first names) expresses this a thousand times better than I did in my initial post. When there is a shift of the demand for money, this will affect all transactions, not only those on final goods. Thus, my first point: gross domestic product is not necessarily the best for monetary transaction.
In fact, as an economist who decided to spend his life doing economic history, I do not like gross domestic product for measuring living standards as well (I’ll do a post on this when I get my ideas on secular stagnation better organized). Its just the “least terrible tool”. However, is it the “least terrible” for monetary policy guidance?
My answer is “no” and thus my proposition to shift to gross output or a measure of “total spending”. Now, for the purposes of discussion, let’s see what the “ideal” statistic for “total spending” would be. To illustrate this, let’s take the case of a change in the supply of money (I would prefer using a case with the demand for money, but for blogging purposes, its easier to go with supply)
Now unless there is a helicopter drop*, changes in the money supply generate changes in relative prices and thus the pattern (and level) of production changes too. Where this occurs depends on the entry point of the increased stock of money. The entry point could be in sectors producing intermediary goods or it could closer to the final point of sale. The closer it is to the point of sale, the better NGDP becomes as a measure of total spending. The further it is, the more NGDP wavers in its efficiency at any given time. This is because, in the long-run, NGDP should follow the same trend at any measure of total spending but it would not do so in the very short-run. If monetary policy (or sometimes regulatory changes affecting bank behavior “cough Dodd-Frank cough”) causes an increase in the production of intermediary goods, the movements the perfect measure of total spending would be temporarily divorced from the movements of NGDP. As a result, we need something that captures all transaction. And in a way, we do have such a statistic: input-output tables. Developed by the vastly underrated (and still misunderstood in my opinion) Wassily Leontief, input-output tables are the basis of any measurement of national income you will see out there. Basically, they are matrixes of all “trades” (inputs and outputs) between industries. What this means is that input-output tables are tables of all transactions. That would be the ideal measure of total spending. Sadly, these tables are not produced regularly (in Canada, I believe there are produced every five years). Their utility would be amazing: not only would we capture all spending (which is the goal of a NGDP target), but we could capture the transmission mechanism of monetary policy and see how certain monetary decisions could be affecting relative prices.** If input-output tables could be produced on a quarterly-basis, it would be the amazing (but mind-bogglingly complex for statistical agencies).
The closest thing, at present, to this ideal measure is gross output. It is the only quarterly statistic of gross output (one way to calculate total spending) that exists out there. The closest things are annual datasets. Yet, even gross output is incomplete as a measure of total spending. It does not include wholesale distributors (well, only a part of their activities through value-added). This post from the Cobden Centre in England details an example of this. Mark Skousen in the Journal of Private Enterprise published a piece detailing other statistics that could serve as proxies for “total spending”. One of those is Gross Domestic Expenditures and it is the closest thing to the ideal we would get. Basically, he adds wholesale and retail sales together. He also looks at business receipts from the IRS to see if it conforms (the intuition being that all sales should imitate receipts claimed by businesses). His measure of domestic expenditure is somewhat incomplete for my eyes and further research would be needed. But there is something to be said for Skousen’s point: total nominal spending did drop massively during the recession (see the fall of wholesale, gross output and retail) while NGDP barely moved while, before the recession, total nominal spending did increase much faster than NGDP.
In all cases, I think that it is fair to divide my claim into three parts: a) business cycles are about the deviation from trends in total volume of trades/transactions, thus the core variable of interest is nominal expenditures b) NGDP is not a measure of total nominal spending whose targeting the market monetarist crowd aims to follow; c) since we care about total nominal spending, what we should have is an IO table … every month and d) the imperfect statistics for total spending show that the case made that central banks fueled spending above trend and then failed to compensate in 2008-2009 seems plausible.
Overall, I think that the case for A, B and C are strong, but D is weak…