Interview with Rick Davis of the Consumer Metrics Institute
Introduction: Richard Davis is President of the Consumer Metrics Institute (CMI). At the Institute, Rick measures real-time consumer transactions as an objective indicator of consumer demand and the associated health of the US economy. In this interview, we explore the history behind the government-published numbers and the reasons prompting Rick to devise better ways to measure the state of the economy.
Ilene: Rick, what got you interested in measuring economic numbers?
Rick: I first became frustrated with the current state of economic data after learning about the history of the collection process and the government’s continued reliance on 70 year old concepts. The government began collecting economic data during Franklin Delano Roosevelt’s (FDR) second term, around 1937. There was concern that the recovery from the 1937-1938 recession (i.e., a recession nested within the Great Depression) was stalling. The economy had been improving significantly from early 1933 through 1936 before the wheels came off the recovery in mid-1937. FDR’s administration realized it did not have adequate data to monitor the economy and the administration asked the National Bureau of Economic Research (NBER) to look into this problem. Wesley Clair Mitchell set out to find data that would help FDR’s administration address its concerns about the U.S. economy.
Wesley Clair Mitchell was a once-in-a-generation economic genius when it came to data collection. He collected over 500 interesting data sets measuring items such as sales, employment, railcar loadings--items that would allow him to constantly monitor the health of the economy. Most of these things are still measured, and the numbers have evolved into the core reports put out by the Bureau of Economic Analysis (BEA).
What frustrated me was that the data sets measured by Dr. Mitchell were developed in the 1930s and designed to capture those things that were important to the 1930s economy. They are not geared for today’s economy. Things that mattered in the mid-20th century simply cannot completely describe what is happening in the 2010 economy.
For instance, to find out what was happening in the music industry in 1950, someone could have gone to a neighborhood music store, counted the Doris Day 45’s in the retail bins and asked the staff a variety of questions about their sales. They’d look at railcar loadings data for the movements of the raw materials used in producing records. Railcar loadings would have captured the materials in the Doris Day 45’s first as crude oil supplies, then as vinyl stock, then again as vinyl record blanks, and finally as finished records in transit to the local wholesaler’s warehouses. In 2010, the music industry’s commerce model is different. For example, the inventory is only partly measured in physical goods. Many people buy music online through iTunes. There’s no inventory for things like downloads, no physical materials to transport.
Dr. Mitchell’s procedures for measuring economic activity made sense in the 1930s, but not now. His methods don’t get us in touch with much of what is happening in 2010. Things have changed.
Ilene: So due to the changes, we need more effective methods to accurately measure economic activity?
Rick: Yes, older methods measure what’s happening in production, and that was key to the economy in the past. In the 1930s, if production started moving that was a good sign. During FDR’s day the vast majority of urban jobs were in factories, and therefore the economic welfare of the vast majority of voters depended on factory production. Now the majority of Americans no longer work on an assembly line.
There’s also a timing issue involved when measuring factory production. Consumers generate 70% of today’s economic activity. But factories are far “downstream” economically from those consumers. The economic stimulus provided by consumers starts upstream and flows down to the factories. In the twenty-first, century the pace of consumer commerce has accelerated to the point that the techniques designed for measuring the types of economic activity important in the 1930s are no longer timely enough to accurately reflect the current state of 70% of the economy. The measures used by Dr. Mitchell are now simply too slow to reflect the current pace of changes in the economy.
To be fair, the mid-20th century measurement techniques did measure what was important then, but they simply do not measure everything that’s important now. Consider, for example, the intellectual property that is being sold when a consumer downloads a song from iTunes. The factory production oriented GDP, as measured even today using 70-year old methodologies, can’t begin to model inventory flows for internet based intellectual property commerce — simply because there are no inventories. The GDP fails to capture much of the commerce going on today at places like E-Bay or Craigslist.
Gross Domestic Product
Ilene: Are there other problems with the GDP?
Rick: Another frustration I have is that the GDP is reported quarterly, and often revised significantly later. I understand that from an academic standpoint, it’s important to revise the number to get it right. The problem is that while the focus in preparing the GDP report is academic, investors need more timely information to guide their investment decisions. Economists can change their numbers; they get do-overs. But investors don’t get do-overs on the investments they made based on incomplete data.
People like me (I’m a physicist) like to measure real-time data accurately. That’s what motivated us [at the Consumer Metrics Institute] to develop the tools we’re using to measure real-time economic data that we believe more accurately reflects the state of today’s economy.
We divide our data on economic transactions into ten segments of economic activity: automotive, entertainment, financial, housing, household, health, recreation, retail, technology and travel. These sub-sections were taken from the National Income and Products Accounts tables used by the BEA. We track these categories of economic activity separately, collecting data from each sub-section of the economy daily. (See http://www.consumerindexes.com/index.html for the latest figures.)
On March 26th the only sub-index that was positive was the automotive index, driven largely by renewed consumer demand for Korean, economy and luxury European cars. On the same day, even the household index, generally one of our most positive indexes, was showing a year-over-year contraction exceeding 2%. The household index includes appliances, apparel and such things as tools, things you can buy at Sears or Walmart, and some of the things you can buy at Best Buy. Best Buy also sells many items that belong to the technology sub-index. Looking at the data, you can see that different sectors of the economy respond differently to downturns in consumer spending. Some spending can be classified as “feel good spending” (trips to the amusement parks, rental cars, small ticket items) and this generally does not decrease during economic hard times as much as “extravagant” spending (big vacations, big ticket items). Little indulgences do better than larger ones during economically troubled times.
Real-time data on what consumers are doing is available from a number of sources. It is used by companies such as Amazon to provide targeted ads to its customers. It’s being used by people in the retail trade, but not by government economists calculating the GDP. So we started to mine that data for macro-economic consumer tendencies using consumer-tracking databases that went back to 2004, For consistency we always look at year-over-year changes in the data, mirroring the annualized growth rates published in the GDP.
We are convinced that the data we’re tracking is accurate. For example, we caught the 2008 recession very well. Over the past several years our data has a lead time of about 17 weeks ahead of the GDP, so we have been predicting future quarters’ GDPs with good accuracy. We are measuring consumer activities so far upstream from the factories that it takes about 17 weeks for the commerce to flow downstream to the government’s production numbers.
Gross Domestic Product Estimates
Ilene: What is your data predicting for the next quarter’s GDP?
Rick: Our data is predicting that the first quarter’s GDP will come in at about 2.5% growth, reflecting our November month end Growth Index numbers. We believe the second quarter’s GDP will show a contraction, at about a negative 1 to 1.5%, reflecting our Growth Index numbers as of the end of February. We are still seeing contraction in that same neighborhood as of the end of March.
In other words, the upstream activity of consumers can be measured in a timely way by using methods that capture electronic real-time transactions. For understanding where the economy is headed, this is better than looking at the numbers that go into the calculation of the GDP, which lag by around 17 weeks. The activity we are measuring now flows downstream to the production segment of the economy over the course of the next three or four months.
We saw the current round of net contraction in consumer demand begin in January, 2010, when the trailing 91 days (number of days in a quarter) turned net negative. The prior peak in consumer activity during this recovery was recorded in August of 2009. Since then, demand has been slowing.
Before that, we saw that recessionary demand bottomed in November of 2008 and rebounded sharply during December, remaining flat for several months after. Consumer demand then surged at the end of March 2009, rising through August 2009 and drifting down since. In other words, during December 2008 and again between March 2009 and August 2009 consumers were spending heavily. In August, the spending spree started to taper off, leading to the current pull back.
Ilene: Why do you think consumers pulled back in August?
Rick: Since we do not know the “why” behind the consumer behavior we monitor, I can only speculate. But I believe consumers turned more cautious when they saw their family members, neighbors and former colleagues remaining without jobs. I believe the lack of improvement in jobs began weighing heavily on the minds of Americans.
Most Americans probably feel that a “jobless recovery” is an oxymoron. When we start seeing people going back to work, I believe spending will start improving again. But now people are saving and trying to pay down debt. Americans are spending about 2% to 4% less now than they were last year. I think this will continue until more people find jobs and get back to work.
Ilene: But aren’t we seeing data on railcar loadings improving lately--why would that be? (For example, see Rail Traffic Surges to New High and Rail Traffic Continues to Rebound, The Pragmatic Capitalist)
Rick: Those reports reflect factory orders. As I mentioned, factory data is way downstream from current consumer activity. They probably reflect inventory buildups or producers incorrectly anticipating a consumer buying spree — something our data is not supporting. We didn’t see consumers buying more durable goods over the last six months. Housing is down in double digits, down around 15% in March, year-over-year. Retail sales are down about 9% year-over-year.
The economy today isn’t in the factories; it is where sales are made – where the consumers are transacting. The railcar loading data is a lagging indicator which measures downstream activity. I believe that railcar loadings are rising because of what consumers were spending in the 3rd and 4th quarters of 2009. Producers may have picked up on the late summer consumer optimism out there and begun rebuilding inventory. But I would caution them to not gear up for a full-fledged recovery yet.
Ilene: If your outlook on the economy is cautious, what does that mean for jobs?
Rick: We’ve seen the wave of unemployment in the private sector — businesses laying off people — but not yet in the public sector. Layoffs and furloughs in the public sector are currently minimal, but I believe they will soon become highly visible. When public sector pain becomes more acute, I believe we’ll see some very public signs of job losses and a general realization that the employment situation isn’t getting better.
The problem facing us is one of visibility. Private sector job losses generally show up as statistics, not headlines. Public sector job losses, on the other hand, are made for journalists. Imagine a school district or municipality laying off 10% of their workforce without it hitting the headlines. It can’t be done. And headlines set the tone for the public’s employment psyche. Americans probably won’t think this recession is over until school districts start a new round of hiring.
Stock Market Disconnect
Ilene: Given that you are not seeing real-time economic improvement, why do you think the stock market going up non-stop?
Rick: The recent stock market move is unfathomable not only from a consumer demand perspective, but from the real world experiences of most Americans. I think most people who are paying attention see the disconnect between the stock market and the real economy, and are not buying into the recovery spin. Frankly there’s a major disconnect between the stock market, both in our data and in the gut feelings of the public.
Ilene: Does that perplex you?
Rick: Yes. But I’ve learned that sometimes you can’t reconcile the stock market’s behavior to what you’re actually seeing in the economy. Americans are not idiots. They know what’s happening. Anecdotally, for instance, when going out to dinner, most Americans will see the waiting staff at restaurants getting older – even baby boomers are increasingly seen waiting tables. These folks are competing with high school and college students for these entry level jobs. Older people are needing new sources of income and becoming less choosy. Perhaps they’ve lost their previous job, perhaps they see they cannot retire. Or maybe they can’t find a job in the field they’re educated to work in, so they displace kids in the lowest level jobs. In any event, this is something people notice. A “jobless recovery” is more than an oxymoron, it’s a spin-doctor’s cure for bad press.
Ilene: That’s what I’ve observed as well, and I don’t see this ending until we address real problems in the economy.
Rick: I agree. But bottom line, I still don’t know what’s driving the market up. Second tier banks, for example, are hurting, not lending. Those banks didn’t get the TARP money. And I think Ben Bernanke might feel like you do when you’re driving along and you suddenly find yourself on black ice and the rear end of the car starts moving in a direction you didn’t intend. You get that sick feeling in the pit of your stomach as you try to get the car in control. Ben’s been “printing money” like crazy (“creating credit” to be more accurate), and lowering interest rates, but the economy is not responding. He’s swerving on black ice and none of his controls are able to make good things happen any more.
Bernanke’s “Printing Press”/Credit Creation Machine
Ilene: You mentioned that the Fed is not really printing money but rather creating credit. Can you explain that?
Rick: Bernanke’s ability to “print money” is an analogy, not to be taken literally. In fact, Bernanke is really expanding the money supply by creating credit. When Ben cranks up the “printing press” he’s really only inviting people to take on new debt.
That is where the analogy fails: if the Fed was actually printing pallet loads of nice new crisp $100 bills, everyone would stand in line for one of those pallets. Everybody sees getting a nice new $100 bill as a good thing. But merely having the opportunity to take on $100 of new debt is not the same. And right now, Ben’s finding out that most people are not interested in taking on new debt.
A better simple analogy for the current situation would be this: the Fed is printing a warehouse full of $100 bills, but unfortunately each bill is covered with a deadly toxin that won’t go away until the economy really recovers. The warehouse is filling up, but nobody’s driving away with truckloads of the toxic bills.
Creating credit requires a willing (and credit worthy) borrower to effectively increase the money supply. Ben’s “printing presses” are currently broken. He’s offering debt, but people aren’t willing to borrow, even with low interest rates. And banks are hoarding money rather than lending it (possibly as a cushion for a coming round of commercial real estate woes). Most people instinctively understand that borrowing more money is not very smart, and banks are loath to repeat the lending errors of the past few years.
So, with banks not lending and people not borrowing, Ben’s tools can’t work. Americans are de-leveraging. The money multiplier effect is acting in reverse, shrinking the money supply as people pay off debt. In real estate, we have had deflation for the past few years. If this deflation spreads to other segments of the economy, the situation will look similar to that in 1929.
Ilene: Why does 1929 remind you of our current situation?
Rick: Actually, one of the things I find most striking about two situations is the type of people involved. In both cases we have career economists who are known for their encyclopedic familiarity with economic history and their skill in using the tools at their disposal. An old saying best sums up my fear: “The generals are always thoroughly prepared to fight the previous war.”
The causes of the Great Depression are many and varied, but the policies of career economists in the Federal Reserve certainly contributed to making a bad situation worse. In 1929, the Federal Reserve had the training, weapons and obsessions to battle the wrong two enemies: hyperinflation and the types of asset bubbles caused by federal deficit spending. They remembered the hyperinflation of the Weimar Republic and were intent not to repeat that. They also remembered the post World War I recession caused by the withdrawal of the fiscal stimulus provided by wartime deficits, and they became obsessed with balancing federal budgets.
To combat hyperinflation, the Fed had tied the M0 money supply to the gold reserves held by the Treasury. This prevented hyperinflation during the 1920s, but by 1929, it had reached the point that all the bullion reserves were fully committed to the existing money supply. When it came time to do battle with catastrophically de-leveraging bank failures and deflation, the Fed’s hands were tied by an inability to combat the shrinking M1 & M2 money supplies.
In that era, over a third of all banks failed, and no bank was considered “Too Big To Fail.” No routine mechanisms existed to intervene when bank failure was imminent. As a consequence, the Federal Reserve Chairman, Roy Archibald Young, could only sit on the sidelines as a run on banks shrank M1 & M2. A deflationary spiral ensued, with surviving banks losing capital and many loans being foreclosed. Because there was no FDIC, depositors at failed banks also lost money, further shrinking M2. The money supply contracted by about a third between 1929 and bottom of Depression in 1933.
The career economists at the 1929 Federal Reserve also felt there was no painless “end game” to the stimulus provided by federal deficit spending. Unless the stimulus could trigger real increases in productivity, no real economic growth would result. When the artificial stimulus was withdrawn, the economy would again crash. They had witnessed this cycle with their own eyes at the end of World War I. For this reason, a Keynesian approach to monetary policy was simply unthinkable.
In 1929, the Fed was equipped and knew how to fight the previous battles, but it was ill-equipped to fight the war it was engaged in. It was prepared to fight hyperinflation and over-stimulation of the economy. It had the ultimate weapon to fight the 1921-1923 hyperinflation that devastated Germany, a currency fully backed by gold bullion. That’s what the members of the Fed knew about and what they were trying to prevent, but those weapons failed miserably in the battle against crippling deflation and a shrinking money supply. The career economists were simply fighting the wrong war.
In 1929, the Fed wasn’t able to do what Ben Bernanke is doing now. For example, this time around AIG was considered “Too Big To Fail.” This was certainly based on cultural history and the bank failure problems of 1929. But the 2008-2010 Fed focuses exclusively on the “To Fail” part of “Too Big to Fail.” It should be focusing on the “Too Big” portion as well. By focusing on “To Fail,” it prevented a bank failure by shifting very toxic assets to the taxpayers, meaning us. The “Too Big” part of the equation simply hasn’t been addressed since someone drove a stake through Glass-Steagall. If AIG hadn’t been “Too Big” I think the Fed could have let it fail just as it lets a half dozen or so other banks fail every Friday, while still protecting the depositors through the FDIC. Why protect the institution? Why save these too big tier 1 banks?
The Fed in 1929 couldn’t dream of doing what it should have done to solve its problems because the career economists running the place couldn’t think “outside the box.” It took a true radical like FDR, with disdain for political and economic convention, to begin to solve the deflation and money supply problems. In 1933, FDR’s administration simply ignored 6,000 years of tradition by banning the private ownership of monetary gold and releasing the dollar from the chains of the bullion reserves. FDR ultimately embraced John Maynard Keynes to some extent, although only the wartime increases in productivity resulted in lasting economic growth.
My fear is that like 1929, the career economists at the current Fed are fighting the wrong war. Historically, a steep “yield curve” created by very low short-term interest rates and a massive infusion of credit have always worked to stimulate the economy. This time around it seems the Fed has hit a stretch of “black ice” and suddenly all the car’s controls are failing. Ben’s been pushing all the buttons, but nothing is creating new jobs and increasing productivity. Maybe everyone’s so busy saving the big banks they can’t see the real enemy.
By our numbers, there still is no sustainable recovery and I feel that the Fed does not have the weapons to fight the real economy’s enemies. Perhaps the real enemies are “Too Big” and the moral hazards created by shifting risks to the taxpayers. Perhaps another one of the real enemies is a sovereign debt level approaching the GDP, a tipping point beyond which Keynesian solutions no longer work. Perhaps the real risk is a lost decade or two, Japanese-style. In any event it will take a great deal of political capital to solve the real problems.
Ilene: What do you mean by political capital?
Rick: I mean that to address the real problems someone needs to take on vested interests with very deep roots in the halls of power. To do that we need a great deal of political capital to expend in fighting an entrenched opponent. Not unlike what FDR had in 1933. I fear there’s not enough political capital or popular support for anything so radical as taking on the big banks, let alone breaking them up. But I will ask anyway: what if we were to disassemble the “too big” banks and split their capital between the smaller tier two banks? Or perhaps end interstate banking as we know it today? Or re-invent Glass-Steagall? Would these measures help prevent the kinds of economic problems we’ve seen over the past four years?
I originally thought that the current Administration had the kind of political capital necessary to take on the vested Wall Street interests. But I’m afraid that the window of opportunity may have passed and that whatever political capital the Administration had when inaugurated has now been spent.
Ilene: What else could be done?
Rick: I would like to see some “outside the box” suggestions for taking on some of the other issues I’ve mentioned earlier. One issue I discussed was the rampant moral hazards created by having taxpayers responsible for loans that no prudent person would ever make. One hundred and ten percent loans to borrowers with undocumented incomes were a bad idea. By shifting transaction risks from the loan initiators to a large pool of investors (and ultimately to the taxpayers) we not only enabled imprudent loan origination, we created new systemic risks. We saw those systemic risks almost take down the entire banking system in 2008. Perhaps some “outside the box” solution could be found that would keep a portion of a loan’s risk permanently attached to the originator’s capital.
Another issue is that our ability to use sovereign debt to solve our economic problems may be reaching the end. Carmen Reinhart’s and Kenneth Rogoff’s wonderful new book "This Time is Different: Eight Centuries of Financial Folly" points out that there is a tipping point for sovereign debt beyond which a country cannot recover without resorting to outright default or (if the country is able to print the currency that the debt is issued in) to default-by-inflation. That tipping point is a sovereign debt to GDP ratio of about 1:1. The United States will cross that threshold within the next few years.
One “outside the box” solution to that problem involves removing the authority to issue new debt from elected politicians, who simply have no collective will power to say “no.” Most politicians have too many reasons to say “yes” to new debt, since doing so avoids hard choices and near term pain to their electorate. The problem, of course, is that there is no painless “end game.”
That may indeed be the greatest weakness of modern democracies, getting politicians involved in economics. In this democracy, some powers are withheld from politicians. For example, they cannot change their terms of their office or alter what constitutes free speech. Perhaps they shouldn’t be allowed to change the levels of our sovereign debt willy-nilly either.
Whatever “outside the box” solutions are needed, rest assured, the career economist/historians at the Federal Reserve and Treasury will be unlikely to pursue them. They will always be fighting the last war using the weapons that worked in the past.
Ilene: Thank you for your perspectives on the economy. Would you like to make a concluding statement?
Rick: Well, I realize that at the moment we can’t change the generals in charge of this war. We probably can’t yet change policies geared for the wrong problems and protecting vested interests.
But we can think “outside the box” ourselves and create a new source of economic data that will challenge the way such data has been collected for the past 70 years. That’s what we’ve tried to do at the Consumer Metrics Institute. We are trying to move economic data into the twenty-first century. Maybe it will encourage people in Washington to do the same.