The Velocity of Crime
If volatility and considerable divergence between individual stocks and sectors has become the norm, it was another "normal" week in the stock market. Blue chips underperformed, with the Dow, S&P500 and Transports largely unchanged. The Morgan Stanley Cyclical index added 1%, while the Morgan Stanley Consumer index declined 1% and the Utilities dropped 3%. Despite today's selling, the broader market enjoyed a strong week. The small cap Russell 2000 added 2%, with the S&P400 Mid-Cap index gaining 1%. Technology stocks rallied sharply, with the Semiconductors surging 9%. The NASDAQ100 and Morgan Stanley High Tech indices gained 3%. The NASDAQ Telecommunications index was unchanged, while The Street.com Internet index declined 2%. Biotech stocks added 2%. Financial stocks were mixed, with earnings warnings from a few key banks and chatter of significant derivative-related trading losses in the marketplace leading the S&P Banking index 2% lower for the week. The Securities Broker/Dealer index added 1%. With gold surging $7 this afternoon, the HUI gold index added 4% this week.
The credit market rally dissipated as the week came to an end, with the yield curve generally steepening. For the week, 2-year yields declined 5 basis points, the 5-year declined 3 basis points, and the key 10-year Treasury yield was largely unchanged. The long-bond saw its yield increase 3 basis points to 5.74%. Mortgage-back yields increased two basis points, while agency yields generally declined 3 basis points. The benchmark 10-year dollar swap spread rose about 2 basis points to 80. The dollar was only marginally higher for the week.
Broad money supply expanded by $17 billion last week, led by a $19 billion increase in institutional money fund assets. Institutional money fund assets have increased $42.4 billion during the past three weeks. Consumer credit increased $14 billion during April, compared to expectations of $8.3 billion. Revolving loans (including credit cards) jumped $9.2 billion, which could have been impacted by the use of credit cards for federal tax payments. Auto (and other personal loans) loans increased $4.8 billion, after a gain of less than one billion during March. According to The Bond Buyer, municipal debt issuance of $27 billion during May was up 70% from last year. Excluding the bonds refinanced, the total amount of new money raised in munis during May was up 59% versus a year ago.
Countrywide Credit this afternoon announced record mortgage-fundings for May, with originations of $10.6 billion more than double year ago originations. At $30.6 billion, it was a record quarter as well, easily surpassing the previous mark of $25 billion. Mortgage refinancing volume was up more than 500% to $6.1 billion. Purchase volume jumped 16% from last year, while home equity volume was up 32%. This week the National Association of Realtors stated that it expects a record year for new home sales this week, with existing sales falling only just short of 1999's record.
Also this afternoon, the Federal Reserve released quarterly flow of funds credit data. While corporate borrowing slowed markedly from a rate of 8.5% during the last year's 4th quarter to 5.1%, the household sector continues to borrow aggressively. Consumer credit (non-mortgage) expanded at a rate of 9.9% (versus 8.1%), while consumer mortgage borrowings slowed slightly to a still strong 7.8% (versus 8.1%). State and local governments increased borrowings to a rate of 7%, compared to 4.8% during the fourth quarter. The financial sector continues to expand aggressively, with outstanding financial sector debt increasing at a rate of 10.6% (versus 13.2%). The government-sponsored enterprises increased borrowings by $88 billion during the quarter (19% annualized), compared to $26 billion (6.6%) during last year's first quarter. Asset-backed securities increased $61.8 billion (13.6%), compared to last year's comparable $33.7 billion (8.3%).
I received the following question this week from a "frustrated economist": "Why would we not expect the huge expansion of money supply not to lead to immediate and substantial inflation? MV = PT. The total volume of goods and services is declining or is at least not increasing. Velocity is relatively stable. Why would prices not go up sharply? Soon?"
Well, I don't know if I can provide a satisfactory answer, but that won't keep me from trying. I think his excellent question touches on important issues that confuse all of us. The purpose of the following discussion is to explore some of the perplexing aspects of contemporary monetary analysis. For one, why the economic community's focus on narrow money and the concept of the "velocity" of these "balances," while at the same time there is a virtual disregard for credit growth?
Let's begin with "velocity." M*V = P*T is known as the "Equation of Exchange," popularized by the great American economist Irvine Fisher. This is an identity with M as the stock of money; V the rate of money velocity (the average number of times during a period that the stock of money "changes hands); P is the average price of transactions; and T is the actual physical volume of transactions. So, the money stock multiplied by the average number of times money changes hands is equal to the price level multiplied by the volume of transactions. This is an equation of all transactions, including finals purchases of goods and services, sales of intermediate goods, and, importantly, capital transactions such as real estate and stock sales. So, with this equation, rising money supply (with velocity held constant) could lead to higher consumer or capital goods (real estate!) prices. While such a transaction-based equation may have had value in the past, in its original context it is difficult to see much relevance today. It was, however, certainly a valuable concept for the development of economics - and monetary theory specifically - as Fisher went on to develop methodologies of quantitative empirical research and the formulation and use of index numbers.
Things got interesting when Fisher's equation was modified and became the famous Quantity Theory of Money, where price level and the level of economic output were linked directly to the stock of money: where M*V = P*Y, with Y being the level of output/income. For this grand jump from total transactions to economic income, V was changed from "velocity of transactions" to "velocity of income." Reading from Dornbusch and Fischer's textbook Macro Economics, "velocity is a useful concept in economic policy making." Well, I guess we will have to take their word for it. I certainly see how the concept of the "velocity" of money has real world relevance in Fisher's work, with his focus on total transactions in an economy where currency was the primary instrument for settling transactions. But does "the velocity of income" pertain to anything remotely real world? How does one go about disaggregating (in the data or conceptually) money changing hands for the purchases that determine an economy's total final output, from money changing hands for all other intermediate and capital transactions? Especially in contemporary society with a sophisticated credit-based payment system and extremely active capital markets. One cannot, of course, and "velocity of income" is basically a plug calculation (income/money stock). Economists maintain a keen interest in velocity for "empirical purposes." More from the textbook: "Thus if we can predict the level of velocity, we can predict the level of nominal income, given the money stock. Further, if velocity were constant, changing the money supply would result in proportionate changes in nominal income." The authors later write, "the concept of velocity is important largely because it is a convenient way of talking about money demand."
From Money and Markets - A Monetarist View, by Beryl Sprinkel: "To convert the 'equation of exchange' into the quantity theory, an assumption must be made concerning independency and dependency of the variables contained in the equation. In other words, the equation of exchange says nothing about which variable provides the motivating force. All it says is that a change in any one variable will be affected by sufficient change in one or more of the other three variables so that MV remains equal to PX. All versions of the quantity theory consider M, the quantity of money, to be the independent variable "
The quantity theory is, of course, the "bread and butter" of the Monetarist school. I have a keen interest in this type of analysis. For one, money certainly "does matter" and I am quite sympathetic to the view that there are significant ramifications for excessive monetary expansion. But at the same time, I am always left with the sense that somewhere along the line monetary theory lost its bearings and veered terribly off course. Having done so, conventional monetary analysis is today only further left in the lurch by the profound changes that have unfolded within the financial system architecture. Theory that emphasizes money's role in exchange is not likely to capture the essence of money as a store of value, the key function I see for money today.
From The Role of Money and Monetary Policy, Karl Brunner (1968): "Three major conclusions (from the Monetarist position) have emerged from the hypotheses put forth. First, monetary impulses are a major factor accounting for variations in output, employment and prices. Second, movements in the money stock are the most reliable measure of the thrust of monetary impulses. Third, the behavior of the monetary authorities dominates movements in the money stock over business cycles."
Traditionally, of course, the banking sector dominated the credit system. Thus the creation of new deposits was the typical residual of credit creation through the bank lending process. With Monetarist orthodoxy holding the Fed in firm control of the monetary system, and with bank lending limited by reserve requirements, economic thinking and textbook explanations came to focus on the idea of a "money multiplier" of reserve balances. This concept had deposits created as banks lent reserves (held by the Federal Reserve) that were then deposited, with lending (less statutory reserves) then creating the next smaller deposit, which would be lent again, and so forth. This explanation creates a misconception, and I would argue it is better explained as a credit creation process, with bank lending creating additional deposits (journal entries) "out of thin air," although constrained by required reserves, if any. This may appear as a subtle distinction, but a critical conceptual point has been missed with the common textbook example: the concept that without the constraint imposed by holding reserves or mandated reserve requirements, the lending process has the distinct capability to create unlimited additional deposits ("infinite multiplier effect").
But as long as the credit mechanism was dominated by the banking industry, my "subtle distinction" was of little relevance, and changes in bank deposits were necessarily indicative of the general credit environment. All the same, one can certainly argue that historically analysts have been too eager to place money as a "cause," instead of appreciating that it is very much an "effect" of lending and credit growth. Generally, this distinction also didn't matter much, with bank loans on the asset side of bank balance sheets matched reasonably well with deposit "money" on the liability side. So, with money and bank credit growth highly correlated, and money data so readily available, why not just ignore lending and credit issues and focus specifically on bank money? Furthermore, if a stable relationship between the stock of money and total lending could be assumed, and that changes in the money stock were highly correlated with output and prices, then wouldn't a relatively stable "velocity" also afford the advantage of a narrow focus on money? Or, looking at this from another angle, with the presumption of a reasonably constant circulation of money balances, there is seemingly no reason to even bother with credit growth. And so it was, with the acceptance of the Quantity of Money Equation and the adoption of a relatively constant "velocity" basically precluding the need to even look at lending or total credit, recognizing instead money as the critical, causal variable.
Well, especially at critical junctures, an appreciation for the accuracy and intricacies of "cause" and "effect" matter a great deal. It is certainly our view that the seminal work in this area - Milton Friedman and Anna Schwartz's A Monetary History of the United States, 1867 - 1960 - erroneously pins responsibility specifically on the Federal Reserve, with its failure to expand the money supply as a cause for the Great Depression. I highlight this analysis, first because it has had a profound impact on the direction of economic analysis and thinking, with particular pertinence throughout this credit bubble environment. Also, it has had a similar effect on the general monetarist framework that has so shaped the curriculums, methodologies, and concepts adhered to by economists, policymakers, and analysts alike.
According to the Monetarist view, with the presumption that money supply is under the strict control of the Fed, it was only necessary for the central bank to have aggressively added reserves to the system to avoid depression. Money supply would have then expanded, and deflationary forces would have been thwarted. Well, at the minimum, this is much too simplistic. There were clearly very complex forces at work, with the decline in the money supply not the "cause" of the Great Depression, but instead symptomatic of the breakdown in the credit mechanism. Sure, money supply was contracting, but wasn't this associated with, or a consequence of, the inevitable impairment to both borrowers and lenders after years of credit and speculative excess? Should analysis have been focused on the simplistic notion of the Fed failing to add reserves, while disregarding the complex issues surrounding a spectacular boom and subsequent bust?
So, disagreeing with Friedman and Schwartz's conclusions, we'll question their methodologies. From an article, The Monetary Interpretation of History, by James Tobin (1965): "What is money? The "money" whose stock F&S (Friedman and Schwartz) trace and explain consists of currency and commercial bank deposits held outside the federal government and the banks. The main questions raised by this definition are these: Why are time and savings deposits in commercial banks, which are not means of payment, included? If they are included, why are similar claims on other financial institutions - notably deposits in mutual savings banks and shares in savings and loans associations - excluded?
On the first question, a decisive practical answer is that it is evidently impossible to distinguish time from demand deposits in commercial banks prior to 1914 F&S recognize that, once the means-of-payment criterion is dropped, drawing the lines that define 'money' is a matter of expediency. What statistical quantity works best? That is, what measure bears the closest and most predictable relationship to measures of economic activity? This is fair enough scientific procedure. But such open-minded pragmatism in the concept and definition of money is an unconvincing prelude to policy conclusions which stress the overriding importance of providing money in precisely the right quantity. Sometimes Friedman and his followers seem to be saying; 'we don't know what money is, but whatever it is, its stock should grow steadily at 3 to 4 percent per year."
Interesting. I am further troubled with the Monetarist doctrine when I read paragraphs like the following (paraphrasing the "New View" critics of Monetarism): "Their (the Monetarists) question relates simply to the behavior of empirical data, not to the behavior of people or markets. How the economic system really works is irrelevant; what Friedman and Schwartz want to know is whether the system works as if its object were to ensure the maintenance of some 'normal relation' between the stock of money and money income To what extent is Friedman & Schwartz's argument directed toward the establishment of causal rather than merely descriptive relations among economic phenomena."
Ok, I think I'm beginning to understand; real world dynamics are subordinate to empirical models. I found another interesting view (The Definition of Money by David Laidler): "The problem of the 'right' definition of money is, in this sense, fundamentally empirical." A leading critic (Don Patinkin) also stated that that the a reason for using income velocity was for "considerations of data availability."
So, it appears, the focus on a narrow definition of money, that remains to this day, was out of expediency (the data was available), and empirical research provided the foundation for this focus on narrow money and Federal Reserve activities, whether narrow money was truly a causal factor or not. There were criticisms (the "New View") and hearty debate that the Monetarist thesis misplaced cause and effect - "that monetary impulses are neither properly measured nor actually transmitted by the money stock. The authors (Gramley, Chase, Kareken, Cacy, Davis, Tobin, Gurley and Shaw) reject the Monetarist thesis that monetary impulses are a chief factor determining variations in economic activity, and they contend that cyclical fluctuations of monetary growth cannot be attributed to the behavior of the Federal Reserve authorities. These fluctuations are claimed to result primarily from the behavior of commercial banks and the public."
It sure looked as if the "New View" was onto something. However, these very real world analyses were cast aside by the tide of a solidifying economic consensus. Quoting Karl Brunner: "The Monetarists' research strategy was concerned quite directly with the construction of empirical theories about the monetary system, whereas the New View indulged, for a lengthy interval, in very general programmatic excursions. Moreover, the New Viewers apparently misconstrued their program as being a meaningful theory about our observable environment." And the rest is history
With confusing concepts, I like to try to create simplified analogies. In a recent commentary, I made reference to a memorable experience from Economics 101. The professor made the point that cause and effect are commonly confused, by noting the statistical correlation between ice cream sales and criminal assaults (with the causal factor being seasonal temperatures). I would like to expand on this line of analysis, attempting to illuminate what I believe is the basis for some critical weaknesses in contemporary monetary doctrine. This analogy includes the local sheriff, having years before become interested in the phenomena of a relationship between ice cream sales and crime. Hoping to lead law enforcement in a new direction, he endeavored to create a model of criminal activity. First, he developed an impressive database including decades of past sales data (easily available from the accounting records of the long-standing ice cream parlor) and extensive detail quantifying the incidence of crime. Sure enough, the data provided a strong statistical correlation, and these results were published with great interest from the law enforcement community. Particular attention was afforded the Sheriff's association of a big increase in ice cream sales to the infamous Great Crime Wave a few decades back. He stated unequivocally that it was the ineptness of previous local government that was fully responsible. After all, if the police would have been properly monitoring ice cream sales and then had reacted appropriately, "we'd have nipped that unfolding crime spree in the bud." Soon, policemen and aspiring officers throughout the community were immersed in the study of this fascinating phenomenon. Ice cream sales were monitored carefully, while some of the most exciting new research focused on shipping data from the dairy farmer. Since the production of ice cream requires cows' milk, the amount of shipments from the diary must certainly be a predictive variable of definite value.
Some, however, weren't so convinced. Quoting one bright skeptic: "Understanding crime is a very complex proposition involving the behavior of individuals, local institutions, labor and housing markets, and the general economy. There are no easy answers, nor simple formulas that will capture the true essence of criminal behavior. In fact, this entire line of methodology is flawed, as ice cream is not causing crime. This fixation on ice cream at the expense of everything else is both misdirected and dangerous." The Sheriff response: "Ok, I've got 90 years of empirical data demonstrating the close correlation between ice cream and crime. Please explain to me what you have? I would like to see your models and the results from testing your variables. What are your correlation coefficients? Furthermore, it is largely irrelevant whether ice cream causes crime or not, as long as we know with confidence that when ice cream sales increase, I put all my deputies on alert and we get ready for a more acts of crime. When sales drop, we can relax, do a bit of community service, and go on vacation. That is all that matters, and in this respect everyone admits my methodology is quite powerful. My motivation is to develop a framework for managing crime, and who can argue that this process is enhanced by closely monitoring ice cream sales?
Such "clear thinking" and a methodology incorporating unimpeachable empirical data were widely appreciated as providing exciting opportunities. It was also a much-welcomed improvement from how things had been done in the past. The townsfolk were ready for a new direction in the battle against crime, and the newfound "professionalism" of its officers and the sophistication and advanced forecasting methods struck a cord. A few, nonetheless, remained incredulous. There were other schools of thought on crime and prevention. One citizen even made a strong case that it was, indeed, the weather that was actually playing a key causal role in any heightened incidence of crime. The Sheriff always reacted uncomfortably to this line of questioning, but was quick to make it very clear that there were several variables and considerable uncertainties to consider when looking at weather patterns. Besides, the local community did not keep historical weather data, and previous attempts at theories incorporating climatic changes proved unsatisfactory. The citizen's request that weather data nonetheless be analyzed carefully going forward went unanswered. The sheriff and his enthusiastic deputies had already built an elaborate apparatus to retrieve and analyze data, from dairy milk shipments, to the number of sugar cones manufactured, and actual scoops sold. Accounts of this new way of policing were greeted with considerable fanfare in law enforcement journals throughout the country. Police academies coast to coast devoted considerable time and effort to teaching all aspects of milk and ice cream production, as well as statistics, icecreamometrics, and mastering the calculation of the Velocity of Crime. Finally, police had something of substance that they could "really hang their hats on." And the townsfolk approved, they had long ago tired of the rising expenditures associated with crime prevention programs. The new focus on ice cream was clearly the way to go.
All went very well for some time. At the first sign of dairies' increasing deliveries to the ice cream parlor, the Sheriff would immediately call up the reservists and get the police force fully prepared for an upturn in criminal activity. The crime rate seemed to respond favorably, much to the delight of the city council, the townsfolk, and the revered Sheriff. But then things started to change. Just as he had done so many times before, the Sheriff dismissed the reservists and cut patrols when ice cream sales began to wane. But for some reason the usual decline in criminal activity did not materialize. While somewhat uneasy, the confident sheriff assumed it was only an anomaly. After checking through all the data carefully (milk deliveries, cones, scoops, foot traffic in and out of the ice cream parlor, etc.) he stood resolutely and comfortably with his theory and policies. Ice cream sales, nonetheless, proceeded to drift lower. The Sheriff and the community waited patiently for the number of crimes to subside, but they rose instead. One of the adept, although "unconventional," scholars of criminal activity noted to the Sheriff that it was a particularly hot and humid summer, and crime always increases in such circumstances. The Sheriff was not interested. After all, he had invested much of his career in developing methodologies and professing the strict relationship between ice cream and crime.
"Ok," exclaimed the citizen in exasperation, "if you will not recognize that temperature changes are a causal factor for criminal activity, at least accept that there are now very close substitutes for ice cream. There's a new shop that has begun selling frozen yogurt, and it certainly appears that this product is going to catch on, increasingly impacting ice cream sales. If you are compelled to stick with your methodologies, at least incorporate that yogurt sales are substitutes for ice cream." The Sheriff's response was polite, while emphasizing that historical data was unavailable to test the empirical relationship between yogurt sales and crime. Besides, with the presumption that yogurt sales would be closely correlated to that of ice cream, current ice cream methodologies remain valid. What was really required was only to adjust for a declining ratio of ice cream to criminal activity - adjust the Quantity Theory of Ice Cream for a change in the "Velocity of Crime". Once factoring the change in "velocity" into the model, the relationship between ice cream and crime will still hold firm.
Well, a few years went by and things were going anything but swimmingly for the Icecreamists. The relationship between ice cream sales and crime had totally broken down. The Sheriff and his deputies studied and tinkered endlessly with The Quantity Theory, but these efforts were fruitless. Since they could determine ice cream sales, prices and the number of crimes, the obvious conclusion was that something very unusual was happening with "velocity." Things, however, took a decided turn for the worst when crime waves began developing immediately after much of the police force would be released for vacations, and with the reservists having long since been terminated. While no one would dare admit it, there was no doubt that the criminals had come to understand the game better than the police. They knew that any steep decline in ice cream sales and the subsequent police "complacency" provided the perfect environment to commit their offenses. The great advantage of opening shops selling inexpensive frozen yogurt did not go unnoticed by the enterprising. With crime rising and the ineffectiveness of law enforcement more conspicuous by the year, the community increasingly lost confidence in these methods of policing. For the criminals, the rising popularity of frozen yogurt and the proliferation of shops was an opportunity of a lifetime. Amazingly, while meticulously monitoring and, in some cases, regulating ice cream sales, law enforcement remained completely disinterested in the frozen yogurt boom.
Curiously, few within the community placed responsibility on the police for the rising level of criminal activity and the occasional intense crimes waves/crises that now seemed to strike intermittently. Law enforcement, no longer capable of explaining the relationship between ice cream and crime, "threw up their hands" and stated that there is simply little they can do - "can't explain it, can't manage it." While he would never state it publicly, the Sheriff did at times ponder how differently Crime Theory would have developed "if all of us would have invested as much time trying to understand the many factors and complexities of criminal behavior and focused on identifying true causal factors rather than on this fixation on ice cream." Though, generally it was "if it weren't for that damn frozen yogurt !"
The greatest mystery, however, is why one of the community's wise elder statesman doesn't step forward and make the case that the study of crime has for too long been terribly misdirected - that it is time to "return to the drawing board," look openly and objectively for alternatives, and begin working toward developing sound theory on the causes and nature of criminal activity. The old models were weak to begin with, and developments over time have made them no longer applicable. It's time to focus on "real world" behavior and relationships, with an emphasis placed on understanding and explaining reality above purely empirical pursuits.
I will be the first to admit that the analogy of the relationship of ice cream and crime to that of money and the economy is extreme. But I do believe it is quite valid to highlight how theory emphasizing empirical expediency over real world phenomena is very much an adventure down a slippery slope. And when ideology and methodologies take precedence over reality in setting policy, one can expect unintended consequences. How did we get to the point where the chairman of the Federal Reserve admits that it is impossible to explain money, let alone manage it? How is it that "credit" became a four-letter word to central bankers and economists, and the explosion of credit created by the government-sponsored enterprises and securitization marketplace is almost completely ignored? Why are economics curriculums still teaching how money is created only by banks lending "reserves" held at the Federal Reserve, when lending through this mechanism is increasingly inconsequential to the runaway money and credit excess instigated by the powerful non-banks completely unconstrained by reserve requirements?
Contemporary financial systems and economies are incredibly complex. Yet, it is my view that a sound theory of money and credit can provide a solid foundation for making sense out of complexity. But it requires good theory. In fact, now more than ever before, the environment beckons for a focus on sound money and credit theory. I certainly don't believe we will be on the road to improved understanding until a new approach is undertaken. Credit cannot remain "a four letter word." There must be a clear focus, and appreciation that new credit creates purchasing power and inflationary manifestations. In analyzing credit, however, we must get away from a narrow focus on "Income" or "CPI" effects. Theory must address that when credit is being created in excess to consummate particular types of transactions (i.e. the explosion of margin debt during the NASDAQ bubble, the surge in junk bond financing for the telecom sector, or the current surge in real estate and consumer debt), there will be various and often complex negative consequences. Credit data should be studied in aggregate, by sector and industry, by instrument, and even by institution when possible. It should be analyzed both from the angle of the borrower and lender.
The analysis of money supply is actually different, but equally as important. Since money is a special type of credit that is created in the lending process, the emphasis should not be as much the impact of money supply growth, as much as should be to understand the lending process responsible for its expansion. Returning to the earlier question, "Why would we not expect the huge expansion of money supply not to lead to immediate and substantial inflation?" First, is it possible to isolate which institution's or sector's lending is most responsible for creating the new liabilities (money)? Today, yes it is. We know both that the GSEs are lending aggressively and that institutional money fund assets are increasing rapidly. This is no coincidence, as a major mortgage-refinancing boom affords the GSEs the ability to expand borrowing in the financial markets to fund the massive accumulation of mortgages that were previously held by other financial institutions - GSE "reliquefication." It does appear that a portion of the current rapid money supply expansion is part of the GSE "conversion" of previous non-monetary debt to "money" liabilities. So, with such analysis, we can garner valuable insights such as the expectation of continued asset inflation in the real estate and credit markets, as well as relatively strong consumer spending. We would expect trade deficits and the continued accumulation of foreign liabilities to be the most pronounced manifestations of this type of credit inflation.
Importantly, monetary theory must also focus specifically on the critical issues of financial stability and fragility. Here, the function of money as a store of value reins supreme. In contemporary financial systems, significant money creation is the output of the process of the financial sector intermediating risky loans into "safe" money. This phenomenon becomes a particularly critical aspect of analysis during asset bubbles, whereby the monetization of rising asset prices evolves into a self-reinforcing bubble. Surging money supply is a residual ("monetization") of this asset inflation process. While it is very difficult (impossible?) to forecast the degree or variation of inflationary consequence(s) emanating from an asset bubble environment, there is absolutely no question as to the accumulating risk for the financial sector as it intermediates escalating quantities of increasingly risky credits. It is certainly my view that there is risk of a negative (CPI) inflation surprise with the continued explosion of money supply, but the much greater risk lies with acute financial fragility. Unfortunately, there is little in current theory that addresses this critical issue.