- Christopher Matthews' case for "Redeeming Greenspan" is based on a flawed understanding of monetary policy.
- The Greenspan Fed aggressively lowered interest rates in the early 00's in spite of surging housing prices, strong loan growth, and high money supply growth.
- "Cheap money" policies by the Greenspan Fed altered the economics of housing, incentivizing both banks and borrowers to engage in reckless behavior.
- The Greenspan Fed deserves the lion's share of the blame for the housing bubble and subsequent financial crisis.
As a frequent "contrarian", I realize that it takes courage to take an unpopular position. When you go against the grain, you will frequently either look brilliant or like a complete fool. CNN Money contributor Christopher Matthews deserves some props for taking on popular opinion in regards to Alan Greenspan in his recent article, "Redeeming Greenspan: Don't Blame the Fed for Bubbles." Unfortunately, the logic behind Matthews' piece is severely flawed.
The thrust of Matthews' argument revolves around a paper by the National Bureau of Economic Research. Researchers examined stock prices after increases in short-term interest rates by the Fed. It was concluded that stock prices tended to rise in the long-term after such movements. Matthews correctly reasons that these increases in short-term rates often occur when the economy is improving and inflation is rising. Hence, one should expect stock prices to rise. Matthews goes on to argue that likewise, "cheaper money" wouldn't necessarily improve stock prices, either. For these reasons, he argues the Fed wasn't to blame for the housing bubble.
Matthews' flaw of reasoning is subtle, yet astronomically large. He makes several great points on Fed policy, and then uses these points to take a logical leap the size of the Grand Canyon. If the Fed raises interest rates based on rising economic demand, then it is "tightening policy"; this much is true. Yet, "tightening policy" is not the same as "tight policy." Conversely, lowering rates may be "loosening policy", but this does not automatically equate to "loose policy." Indeed, with rapidly changing market conditions, the Fed could run "loose policy" or "tight policy" via inaction rather than through rate changes.
In other words, we can't assess the "tightness" of Fed policy merely by examining Fed rate changes. We have to look at the market and compare what's happening in the outside world to Fed policy. Taken to its logical extreme, Matthews is arguing that monetary policy doesn't matter, and that raising the Federal Funds Rate to 20% would have no impact at all. This argument does not stand up to scrutiny. Moreover, it would seem to be in opposition to the most seminal economic work of the 20th Century: Milton Friedman and Anna Schwartz's A Monetary History of the United States.
Why Greenspan and the Fed are to Blame
Examining the market conditions of the 00's, one can conclude that Greenspan set rates well below market levels and ran very loose monetary policies. How do we know this? There's not one straight-forward metric, but we can piece together several pieces of evidence supporting this thesis, including surging housing prices, rapid loan growth, double-digit increases in money supply, and Fed policies that focused on CPI and PCE, rather than asset prices.
I've addressed the latter issue in numerous articles, but I'll rehash it again here. In 1983, the Bureau of Labor Statistics removed housing price data from the Consumer Price Index. While the move likely seemed less than monumental at the time, it greatly altered the way that Fed policymakers viewed inflation. Fed officials, and many economists, assume that CPI, alongside its sister stat, Personal Consumption Expenditures ["PCE"], are the best proxies for inflation.
The problem is that inflation, often driven by surges in bank lending, often manifests itself in asset prices first. The logic behind this is simple: banks typically lend against hard assets that have more definitive values. If the borrower defaults, the bank then has something to re-possess to make itself whole. As a result, a surge in lending is most like to impact asset prices first. Only later does this inflation flow down to other items such as rent, utilities, and goods. Hence, when faced with a lending surge, CPI and PCE tend to be lagging indicators of inflation.
When the BLS removed housing prices from the CPI, it was replaced with Owners' Equivalent Rent, an odd concept that contrasts dramatically with housing prices. While housing prices tend to be highly volatile, OER has barely budged out of its 2% - 4% range over the past 20 years. Many economists reason that this makes CPI more stable and easier to interpret, but it also makes it more detached from market realities.
If we were to re-engineer the CPI to include housing prices, rather than owners' equivalent rent, the perspective changes dramatically during the housing bubble years. As you can see below, housing-adjusted CPI ("Alt-CPI") was persistently and dramatically higher than official CPI from 1998 - 2005. Looking at CPI and PCE, Fed policymakers opined that inflation was low, but one look at skyrocketing housing prices tells another story.
(click to enlarge) As a result of this, there was a huge disconnect between Fed policies and market realities. The Fed, under Greenspan, was actively lowering rates, while housing prices were shooting through the roof. The Fed didn't start to catch up till around 2004, at which point, the bubble was already too huge to pop without major ramifications in the economy.
Monetary Policy Rules
Even if we ignore the surging housing prices, and assume that CPI was an appropriate proxy for inflation, Greenspan's policies still look bad. The Taylor Rule is a commonly used guide for the Fed's short-term interest rate policy. It was developed by Stanford economist John Taylor. Using the official CPI numbers for the Taylor Rule, it suggests "loose policy" in the early 00's.
The chart below compares the Taylor Rule and the Federal Funds Rate. You can see that the two tend to follow one another fairly closely, but veer apart in the early 00's.
A more useful way to analyze this chart is to subtract the Taylor Rule recommendation from the Federal Funds Rate to see the "gap" or difference between the two. Using this perspective, we can see from about 1999 through 2006, the Taylor Rule suggested that Fed monetary policy was too loose.
We should note that the Taylor Rule is far from perfect, and it should be expected that the short-term interest rates might veer away from its recommendation for short periods of time. However, you can see that from 1998 till around 2005, the Taylor Rule recommendation consistently favored tighter monetary policies.
Remember, that the Taylor Rule uses the CPI figures that we've already established might be flawed as proxies for inflation. The Taylor Rule - Fed policy disconnect becomes much greater if we plug in the housing-adjusted CPI numbers. Under this prism, Fed monetary policies look extremely loose for most of the period running from 1998 - 2005.
If the Taylor Rule has any merit, then it becomes even clearer that the Greenspan Fed ran policies that could be considered "loose" and "recklessly loose" if we begin to factor in the persistent double-digit housing price increases that CPI misses.
If the housing price surge and the Taylor Rule weren't enough evidence to suggest that the Greenspan Fed enacted too loose monetary policies, then we can also take a look at lending and money growth. Below, we examine bank credit growth alongside the Federal Funds Rate. The fact that bank credit grew rapidly in the early 00's was not cause for alarm in and of itself. Indeed, there have been several similar rises in the past 70 years. Rather, it's the monetary policy reaction to this from the Greenspan Fed: lower short-term interest rates.
On one hand, we might conclude it's understandable that the Fed was concerned about a rapid plunge in bank credit growth in the in 2001. On the other hand, bank lending growth never fell to critical levels and rapidly recovered; at which point, the Fed lowered interest rates some more. The Fed's general position at this time period seems to have been: "If the economy looks weaker, lower rates. If the economy looks stronger, also lower rates."
In my last article, I took a look at Divisia M4 money supply, a metric recommended by Economist Steve Hanke to examine money supply. M4 is the broadest measure of money and "Divisia M4" adjusts M4 for opportunity costs. Once again, this metric proves useful and appears to show very rapid money growth during the early 00's.
Divisia M4 reached double-digit levels, hitting 12.0% year-over-year growth by November 2001. Surprisingly, this constitutes the highest growth in the entire series, dating back to 1968.
The Fed's flawed response becomes particularly apparent when we examine Divisia M4, alongside the Federal Funds Rate. Here, we can see that the Fed lowered short-term interest rates from 6.5% in December 2000, all the way down to 1.75% by December 2001. When looking at the data, this appears to be a dramatic over-reaction to the recession caused by the collapse of the tech bubble, as Divisia M4 money growth never dipped below 5%. While we only have about 37 years of data, the average growth in that timeframe is 5.72%; suggesting that money growth in late 2000 was only slightly below-average.
All of this provides more evidence that Fed policy was simply out of touch with market realities. Almost every piece of data here suggests that (NYSE:A) asset prices were surging, (NYSE:B) loan growth was increasing rapidly, (NYSE:C) money growth was above-average, and (NYSE:D) the economy was in OK shape overall. Yet, the Greenspan Fed was so desperate to avoid a depression after the 2000 tech crash, that it completely ignored these trends. Fed policymakers justified low-rate policies by focusing on the "low inflation" shown in CPI and ignoring asset prices, loan growth, and money supply.
The conventional wisdom is that there is a lot of blame to go around for the '08 Financial Crisis. It's true that there were many irresponsible bankers, irrational homebuyers, and poor policies from the Federal government. Yet, the Federal Reserve Bank has an outsized influence on the US economy, far exceeding that of any other institution or individual, even including the "bulge bracket" banks. The Greenspan Fed bares the lion's share of the blame for the financial crisis, as the "reckless actions" by large banks and homebuyers were only empowered via cheap money provided by the Federal Reserve Bank.
By examining the data in the late 90's and early 00's, we can see that US housing prices were surging, lending was growing rapidly, money supply was at above-average levels, and the Federal Reserve's response was to aggressively lower interest rates. While the tech bubble collapse weakened the economy somewhat, the Fed reaction was excessive, and even after it was clear a recovery was underway, the Greenspan Fed continued to lower rates further.
Unfortunately, there's little evidence that we have learned the right lessons from the downturn. Instead of focusing on more prudent monetary policies, we've continued to run loose policies, but have now coupled them with tighter regulatory policies that hamper economic growth. There is still "cheap money"; it's just not available to middle income homebuyers any more. The likely impact is that the next bubble won't be in US housing; it will merely shift to some other asset class.
By the way, we now have record margin debt, corporate debt, and student loan debt as a percentage of output. Meanwhile, many emerging markets appear to have record levels of debt, as well. I make no claims to see the future, but it's difficult to examine the current environment and not see a lot of "risk" being created.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.