US equity valuations are currently not symptomatic of a stock market bubble. However, with frenetically rising stock prices in the face of simultaneously deteriorating macro and micro fundamentals, two questions are begged: First, what is driving the rise in stock prices represented in broad indices such as the S&P 500, Dow Jones Industrial Average and Nasdaq? It is clearly not the recent evolution of fundamentals in the form of macroeconomic and corporate earnings forecasts. Second, how sustainable is the current rally and how far can it go?
The present article is about potential developments in the intermediate term time frame, which I define as 0-12 months. Specifically, in this essay, I will argue that many of the requirements for the formation of a liquidity-driven stock market bubble are currently falling into place.
The Story of Excess Liquidity And Normalization of Liquidity Preference
Below I present a chart of a metric that I call "Systemic Liquidity." This metric measures most of the major readily available sources of liquidity for US households and businesses as a percent of total nominal income or GDP. This measure of systemic liquidity includes the following:
1. Currency outside the U.S. Treasury, Federal Reserve Banks, and the vaults of depository institutions (essentially, US Dollars in circulation or the cash households and businesses physically possess).
2. Traveler's checks of nonbank issuers.
3. Demand deposits (i.e. checking accounts).
4. Other checkable deposits (OCDs), which consist primarily of negotiable order of withdrawal (NOW) accounts at depository institutions and credit union share draft accounts.
5. Savings deposits (which include money market deposit accounts, or MMDAs).
6. Small-denomination time deposits (time deposits in amounts of less than $100,000).
7. Balances in retail money market mutual funds (MMMFs).
8. Institutional money funds.
I divide this aggregate measure of liquidity by nominal GDP in order to get a sense of "systemic liquidity," or the total amount of liquidity that is currently available to US households and businesses. The black trend line is an estimate of a "normalized" level of systemic liquidity, derived by linear regression.
There are two important things to observe in this graphic. The first thing to note is that from 1959 until 2000, US systemic liquidity as a percent of GDP behaved in a cyclical and mean-reverting fashion, just as standard economic theory would predict: Liquidity would tend to rise during recessions and during the initial phase of recovery as households and businesses demanded increased liquidity for precautionary reasons and the Fed accommodated this increased demand. As the economic cycle progressed, liquidity as a percent of GDP would then contract as liquidity preference declined and the Fed withdrew excess liquidity from the system. This mean-reverting behavior is symptomatic of a well-functioning flexible monetary system.
The second thing to observe in this graphic is that systemic liquidity as a percent of GDP was in a discernible secular downtrend from 1959 until the events of 2000-2003. This secular decline in systemic liquidity as a percent of GDP makes perfect sense as financial innovations over time enabled households and institutions to hold less and less conventional sources of liquidity as a percent of their incomes.
For example, at the household level, financial innovations such as credit cards, and various financial facilities designed to enable individuals to easily tap into their home, pension and insurance equity greatly facilitated access to liquidity. This greater ability to access liquidity meant that over time, the need for households to hold conventional sources of liquidity (such as cash, demand deposits, savings deposits and etc.) for precautionary and transactional purposes declined as a percentage of their incomes.
Similarly, in the business sector, financial innovations in money markets, securitization, swaps and derivatives enabled corporations to access liquidity with increasing ease. As a consequence, all things being equal, US businesses were enabled to reduce their holdings of traditional sources of liquidity for precautionary and transactional purposes as a percent of their incomes.
2000-2003 proved to be a key turning point in US economic history, and the evolution of systemic liquidity in particular. A succession of events including the bursting of the US Internet bubble, the 9-11 terrorist attacks and the Afghan and Iraq wars caused a spike in the demand for liquidity, which the Fed accommodated on an until-then unprecedented scale.
The ramping up of systemic liquidity during this 2000-2003 crisis period might not have represented such a major problem if the Fed had not been excessively slow in withdrawing its accommodative monetary stance once economic and financial conditions were on the mend. However, for various reasons, the Fed never fully engineered a true normalization of systemic liquidity levels. As a result, systemic liquidity was still at around 60% of GDP at the end of 2006, versus an indicated level of 55% or below which history suggests would be the appropriate level at that advanced stage of the economic cycle.
Therefore, while trending down as a percent of GDP, systemic liquidity remained at extremely high levels throughout the mid 2000s and this helped to fuel the creation of asset bubbles - particularly in the housing and mortgage securitization sectors.
The bursting of these bubbles precipitated a massive and systemic financial system crisis in which the demand for liquidity spiked to monumental heights. In response, the Fed accommodated this spike in the demand for liquidity via an aggressive policy of quantitative easing (QE) on an absolutely unprecedented scale. The result was an explosion in systemic liquidity available in the US economy from about 60% of GDP at the beginning of 2007 to almost 80% at its peak in 2009.
Again, one could very reasonably argue (as I did at the time) that the Fed's accommodation of the spike in the demand for liquidity during the 2007-2009 financial crisis was justified given the high demand for liquidity, and that this did not necessarily represent a major threat in and of itself in terms of fueling consumer or asset price inflation. Liquidity will not cause price inflation (consumer or asset) if it is merely hoarded for precautionary reasons. However, just as occurred in the 2003-2006 period, the problem has presently been that the Fed has been excessively slow in engineering a normalization of systemic liquidity levels.
The Story of Asset Bubbles
I want to be clear: Elevated or "excess" levels of systemic liquidity will not in and of itself cause stock prices to rise, nor much less cause a stock market bubble. Automatically making this association is a mistake that many people make. For example, stocks rose steadily during the 1990s at the same time that systemic liquidity was sharply trending downwards. Similarly, from 2007-2009 stock prices collapsed despite an unprecedented explosion of liquidity.
Many people carry around the false mental model that if the quantity of money or liquidity rises and the quantity of goods (including investment goods) remains constant, then prices will rise. This is simply false.
It is extremely important to understand that liquidity will not cause the prices of goods (investment and consumer) to rise if it is hoarded by households and businesses rather than spent. Liquidity will only affect prices insofar as there is a change in the marginal preferences by households and/or businesses to possess liquidity relative to possessing other types of goods (including investment goods such as stocks). In particular, the availability of liquidity (in any quantity) will generally only affect the prices of consumer and/or investment goods to the extent that "liquidity preference" changes (more on this below).
If liquidity preference rises, this can have a disinflationary or deflationary impact. If liquidity preference falls, this can have an inflationary effect. If liquidity levels are high and liquidity preference remains the same, there will generally be no pressure on prices. If liquidity is high and liquidity preference actually increases, deflation can occur. Therefore, it is important to understand that there is no automatic link between the quantity of available liquidity and stock prices.
That is not to say that excess liquidity is irrelevant. Excessive systemic liquidity can become highly relevant under certain circumstances. Excess liquidity can enable consumer and/or asset price inflation if and when liquidity preferences decline and the excess liquidity is not withdrawn from the system in a timely fashion.
So why do I say that the current extraordinarily high levels of systemic liquidity at this particular stage of the business cycle poses serious risks to the economy? To understand this, we must first fully understand the concept of "liquidity preference." Liquidity preference refers to the desire of households and businesses to hold liquidity. Liquidity preference tends to behave in a mean-reverting fashion driven largely by fluctuations in risk aversion.
For example, when the economy gets worse the demand for liquidity by households and businesses tends to rise for precautionary reasons and because access to credit becomes scarce. Likewise, as the economy stabilizes and improves, the demand for liquidity by households and businesses tends to decline as the precautionary motive dissipates and access to credit expands.
The reason that declining liquidity preference in the context of extraordinarily high systemic liquidity can be dangerous is that as households and businesses gradually shed their "bunker mentality," they will no longer want to hold so much liquidity. In particular, they will not want to hoard so much of their net worth in liquid assets at wealth-destroying negative real rates of interest. This implies that as households and businesses become less fearful and risk-averse, they will increasingly deploy their accumulated cash hoards towards increasingly discretionary consumption and increasingly risky investments. This can fuel price inflation and/or relative price distortions in consumer and/or investment goods.
Unfortunately, just as occurred during the 2003-2006 period, the current failure on the part of US monetary authorities to normalize systemic liquidity levels as economic and financial conditions normalize may very well result in the formation of new asset bubbles. Indeed, the situation today is much more worrisome than it was in the mid 2000s. Presently, systemic liquidity is well above 75% of GDP, versus a normalized level that should probably be around 55%. This implies that "excess liquidity" in the US economy in the order of 20% of GDP. This is an utterly unprecedented and dangerous level of systemic liquidity in the context of an economy that has stabilized and is now progressing through an intermediate stage of the business cycle.
In the follow-up article in this series, I will address the issue of whether high unemployment makes the currently extraordinary levels of systemic liquidity in the context of declining liquidity preferences any less dangerous to the economy. I will touch on the subject of relative price shifts amongst and within consumer and investment goods. Finally, I will show that given the low real yields on fixed income instruments, a very large portion of the excess systemic liquidity that is currently in the US economy will probably be deployed to equity investments and stock market investments in particular, via ETFs such as S&P 500 SPDR (NYSEARCA:SPY), SPDR Dow Jones Industrial Average (NYSEARCA:DIA) and PowerShares (NASDAQ:QQQ), potentially causing a dangerous bubble.