A sectoral analysis of the financial instability hypothesis

A sectoral analysis of the financial instability hypothesis

The Quarterly Review of Economics and Finance 53 (2013) 450–459 Contents lists available at ScienceDirect The Quarterly Review of Economics and Fina...

3MB Sizes 0 Downloads 44 Views

The Quarterly Review of Economics and Finance 53 (2013) 450–459

Contents lists available at ScienceDirect

The Quarterly Review of Economics and Finance journal homepage: www.elsevier.com/locate/qref

A sectoral analysis of the financial instability hypothesis Robert F. Mulligan ∗ Western Carolina University, United States

a r t i c l e

i n f o

Article history: Received 8 December 2011 Received in revised form 20 May 2012 Accepted 27 May 2013 Available online 6 June 2013 Keywords: Hedge Speculative Ponzi finance units Debt depreciation Malinvestment liquidation Overconsumption

a b s t r a c t Hyman Minsky’s Financial Instability Hypothesis (FIH) is applied to various North American Industrial Classification System (NAICS) industry groups, and it is found that some sectors develop much more closely in accordance with the FIH than others. Minsky categorized firms based on the relationship between cash flow and debt service requirements: hedge finance units, whose operating revenues are adequate to service current interest and principal on their debt; speculative finance units, which can meet interest payments but cannot pay down principal; and Ponzi finance units, which cannot meet current interest payments. The FIH is related to, as well as supportive of, Austrian Business Cycle (ABC) theory, because interest rates are negatively correlated with the proportion and market value of speculative firms in several sectors. © 2013 The Board of Trustees of the University of Illinois. Published by Elsevier B.V. All rights reserved.

“He who refuses nothing,. . .will soon have nothing to refuse.” Martial, Epigrams XII, 79 1. Introduction Minsky (1975, 1982, 1986, 1992), classifies heterogeneous firms in three categories, according to the relationship between available cash flow and debt service needs—(1) hedge finance units generate sufficient cash flow to service both interest and principal, reflecting the discipline not to borrow more than their cash flow justifies; (2) speculative finance units borrow so much their cash flow covers interest payments but not repayment of principal; and (3) Ponzi finance units borrow so much they cannot even cover interest on their debt. In Minsky’s Financial Instability Hypothesis (FIH), protracted periods of prosperity lead endogenously either to progressive acceptance of greater risk on the part of firms, or a mistaken under-evaluation of the market risk to which firms are exposed. As the expansion phase of the business cycle wears on, ever more-leveraged firms expose the financial sector to greater risk, thus actual risk-adjusted returns are lowered economy-wide as the economy becomes progressively more dominated by speculative and Ponzi finance units. Firms can increase their degree of overleveraging merely by failing to meet the sales or earnings expectations assumed to justify current borrowing levels, as well as by aggressively pursuing additional leverage.

∗ Tel.: +1 828 227 3329; fax: +1 828 227 7584. E-mail address: [email protected]

Higher degrees of leverage are seen as normal as the expansion phase of the business cycle wears on and more prudent approaches to lending and borrowing are progressively abandoned. Once Ponzi finance units reach a level of indebtedness where they can no longer borrow in increasing amounts based on fixed collateral, they are suddenly forced to sell off assets to make interest payments, and once the economy reaches a critical threshold of Ponzi finance units, this creates an oversupply of assets offered for sale, and the resulting debt deflation causes a financial crisis and liquidity shortage. Because the distribution of FIH categories is highly sensitive to earnings fluctuations, a crisis state can be brought about by a deceptively low critical mass of Ponzi and speculative finance units. Furthermore, since speculative bubbles are localized in particular industries, we would expect to see different degrees of overleverage develop in different sectors. A sufficient degree of speculative and Ponzi finance concentrate in a particular industry may be sufficient to drive a financial crisis affecting most or all other sectors. This paper uses a large 2002–2009 quarterly data set of all publicly traded North American firms and foreign firms traded on North American exchanges, a total of 8707 companies. Eight industry groups are selected for examination, based on the twodigit NAICS prefix: Agriculture, Forestry, Fishing, and Hunting (11), Mining, Quarrying, and Oil and Gas Extraction (21), Utilities (22), Manufacturing (31–33), Transportation and Warehousing (48–49), Information (51), Real Estate, Rental, and Leasing (53), and Professional, Scientific, and Technical Services (54). Financial ratios are used to classify these firms in each quarter according to Minsky’s FIH categories. Market value is used to weight the categories for

1062-9769/$ – see front matter © 2013 The Board of Trustees of the University of Illinois. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.qref.2013.05.010

R.F. Mulligan / The Quarterly Review of Economics and Finance 53 (2013) 450–459

firms in each NAICS sector. Results provide direct empirical support for the FIH on a sector-by-sector basis. The FIH is then reinterpreted in terms of Austrian Business Cycle (ABC) theory, which depends on inflationary credit expansion to drive the unsustainable prosperity. According to Minsky’s FIH, unsustainable prosperity emerges endogenously as long periods of economic expansion make borrowers and lenders alike more willing to engage in activities for which they systematically underestimate the actual risk. Protracted prosperity, stability, and growth naturally leads actors to progressively underestimate risk, driving a wedge between ex ante perceived risk-adjusted returns, which determine borrower and lender behavior, and actual ex risk-adjusted returns. It becomes clear that this process is amplified and exacerbated by credit expansion ABC theory posits as the cause of the business cycle. Minsky’s FIH helps flesh out some of the missing dynamics of the malinvestment liquidation phase of ABC theory, and the two views turn out to be surprisingly complementary. Examining the dynamics of debt financing by sector introduces some of the firm heterogeneity on which both ABC theory and the FIH depend, and which is often disregarded in empirical macroeconomics. The remainder of this paper is structured as follows: Section 2 recapitulates recent empirical findings based on the FIH; Section 3 presents the empirical results of this paper; Section 4 discusses the relationship between the FIH and ABC theory; Section 5 presents concluding comments.

2. Earlier empirical findings A number of studies have refined or elaborated on the FIH (e.g., Dos Santos, 2005; Dos Santos & Zezza, 2008), but empirical studies have been exceedingly rare. Sethi (1992) developed a model of firm behavior based on the FIH, in which information constraints, transactions costs associated with the acquisition and evaluation of relevant new information, and the dynamics of learning, all contribute to the progressive acceptance of over-leverage by both borrowers and lenders. His computer simulations were able to mimic business cycle behavior better than rational expectations models, when the rational expectations forecasts and agent expectations were not publicly observable, but had to be discovered through observation over time. Silipo (2011) examines whether asset portfolios increasing include overvalued risky assets over the course of an expansion. As agents become more optimistic, they become less risk averse, or simply under-assess risk exposure, and assets used as collateral for loans thus become over-valued (Minsky, 1982). As capital asset price volatility falls over the course of an expansion, standard valuation techniques lead agents to conclude that assets are undervalued based on expected cash flows, which have been increasing and may be over-assessed, while cash flow volatility is increasingly likely to be under-assessed as the expansion progresses. If risk exposure is wrongly assessed at too low a level, risk-adjusted returns are misperceived as greater than their true underlying values. Asset valuations are thus systematically biased upward, and to borrowers and lenders caught up in the euphoria of an extended boom, these inflated asset values seemingly justify higher credit limits. And in the absence of realized losses, lenders and borrowers mistakenly conclude that cash flows justify increasingly higher debt levels. As the boom period of steadily growing cash flows extends in duration, memory of past downturns and market volatility recedes, and capital asset prices rise in relation to current cash flows, acceptable debts increase in relation to income streams and collateral requirements (Minsky, 1982, p. 144), the quantity and quality

451

of collateral required by lenders falls, as investment spending and profits increase, all contributing to the transformation of the financial structure from an initial one of transparent and robust simplicity to one of increasing complexity, hidden dependency, concealed weakness, and fragility (Minsky, 1982, p. 111; Silipo, 2011, p. 443). Silipo examines 1991–2009 data, and among his conclusions are “that investor confidence and risk appetite shaped the business cycle more than monetary policy and the cost of borrowing” (2011, p. 447). Such measures as the Goldman Sachs Risk Aversion Index fall steadily over the expansion and rise abruptly at the onset of the 2001 and 2007–2008 crises. In addition, leading up to the financial crisis in late 2007, bank’s asset holdings more than doubled, and the majority of this balance-sheet expansion occurred in increasingly risky assets (Silipo, 2011, pp. 448–449). Minsky (1986) cites securitization as a vector of financial fragility, because avoids the traditional regulatory constraints on growth of liquid assets, bypassing reserve requirements, central bank regulation, and monetary policy. As financial obligations become increasingly layered, the economy becomes more vulnerable to the inevitable liquidity crises which must occur eventually at some location. Highly layered obligations, made possible through extensive unregulated securitization, ensure that any liquidity shortage will spread from sector to sector, given the fragile and interdependent financial structure (Minsky, 1982, p. 132). Silipo observes that expansion-generated portfolio transformations also shrink the financial system’s domain of stability, and thus make liquidity crises more likely. At some point, such a crisis becomes inevitable. One feature of the last recession which does not strictly fit the FIH, is that the liquidity crisis occurred among those households and financial institutions which had run up the highest levels of indebtedness, rather than among the relatively less indebted corporate sector (Silipo, 2011, p. 452). This explains why the crisis occurred in the overextended financial sector. Silipo also observes that interest rates rose along with amounts borrowed, although ceteris paribus, borrowing should fall as the cost of borrowing rises. Unless conventional supply and demand theory can be set aside, this could only happen due to an increase in loan demand (or a decrease in supply, which can clearly be ruled out), and one of the possible causes of such an increase in demand, is an increase in borrower confidence. The flight to quality Minsky (1982, p. 131) predicts for the financial sector once the crisis and debt deflation hit, as a response to greater uncertainty by profit-maximizing, risk-averse banks, clearly occurred (Silipo, 2011, p. 453) as the monetary base was quadrupled, though monetary aggregates only increased by a comparatively moderate factor of about one-third (as of May 2012). Silipo’s study is noteworthy as one of the few empirical examinations of the FIH. Most applications of the FIH have employed computer simulations. Keen (2013) uses simulation studies based on his own FIH model (Keen, 1995) to reproduce the dynamics of the expansion and collapse. In addition to increasing risk and instability over the course of the expansion, Keen also shows that over the expansion phases of the cycles, the wage share of income fell, non-financial business incomes stabilized, and financial-sector income rose. In presenting their model and simulation studies, Chiarella and Di Guilmi (2011) observe that in the stock market, heterogeneous agents form expectations which determine the market valuation of assets, taking advantage of the market’s heterogeneous expectations to proxy firm and capital heterogeneity. In addition, to model the endogeneity of money, they model monetary aggregates as being linked to the total amount of financial assets, rather than a fixed multiple of the monetary base, an approach which

452

R.F. Mulligan / The Quarterly Review of Economics and Finance 53 (2013) 450–459

is supported by the recent monetary literature (e.g., Binner et al., 2010) and recent experience. Their simulations effectively mimic the business cycle over a variety of conditions. They conclude, “the economy can be stabilized by reducing its capacity to create endogenous money and the maximum debt ratio allowed” (p. 1167). Both Keen (1995) and Chiarella and Di Guilmi (2011) incorporate a conventionally negative Phillips curve tradeoff between inflation and unemployment in their FIH models, positing that price-level increases lead to, or perhaps are caused by, increases in employment. A number of recent empirical studies have shown that either this relationship is exactly reversed—inflation actually lowers employment, even in the short-run—or that the intertemporal dynamic is somewhat more complicated, and that increases in the price level may result in moderate, though temporary, increases in employment real economic activity, but result in larger permanent losses of jobs and real output (Niskanen, 2002; Reichel, 2004; Moghaddam & Jenson, 2008; Mulligan, 2011). If the increase in the price level over the course of an unsustainable expansion is either driven by endogeneous increases in demand for liquidity, or indirectly by increases in risk tolerance, the fact that it ultimately results in negative economic growth, is actually supportive of both the FIH and ABC theory.

3. Data and results COMPUSTAT provided data for 8707 firms traded on North American exchanges between 2002 and 2009. Firms in each twodigit industry group were first divided into hedge, speculative, and Ponzi-finance units based on interest coverage (IC), defined as (net income + interest expense)/interest expense. An IC of greater than or equal to 4.00 to was selected to identify hedge finance units, between 4.00 and zero for speculative finance units, and less than zero for Ponzi finance units (Fig. 1). The FIH predicts that the percent of speculative and Ponzi firms should fall during the recovery phase, increase over the course of the business expansion between recessions, and rise again during or prior to the next recession. This inverted-u pattern can be seen to differing extents, for the mining, manufacturing, transportation and warehousing, and information sectors, but is completely absent from agriculture, utilities, real estate, and professional, scientific, and technical services. Formal hypothesis tests are reported in Table 1, revealing that the FIH is borne out for every sector except utilities and services, as evidenced by the significantly positive coefficient on the square of the time index. If, instead of focusing on the percentage of firms in each category, we examine total numbers of listed firms decomposed by FIH category (Fig. 2), we find stronger evidence for Minsky’s theory. The number of firms listed overall grows between recessions, but falls after the start of the 2007–2009 recession, for all sectors except utilities. In mining, manufacturing, and transportation and warehousing, this increase in the number of firms occurs among hedge firms, though for agriculture, Ponzi firms grow most, for real estate speculative firms grow most, and for information and services the growth in firms is spread more evenly across FIH categories. Generally, the number of listed firms grows until the first quarter of 2007, shortly before the onset of the recession, and that the number of Ponzi and hedge firms grows fairly steadily until the recession becomes imminent. When the financial crisis occurred in the fourth quarter of 2008, numerous hedge firms became either speculative or Ponzi, and speculative firms transformed into Ponzis. The impact of the financial crisis can be seen most dramatically for agriculture, mining,

manufacturing, and transportation and warehousing, where the number of Ponzi firms sharply increases around the start of 2008. These patterns become more dramatic when each firm is weighted by its market value. First, we look at the percentage of total market value contributed by all listed firms in each of the three categories (Fig. 3). The data for market value only goes back to the start of 2006. The graph for the agricultural sector is dominated by seasonal financing patterns—and interestingly, the percent of market value attributable to hedge finance units grows steadily even as we enter the recession, if seasonal patterns are removed. The percent of market value of speculative and Ponzi firms increases around the middle of 2008 for mining, manufacturing, transportation and warehousing, information, and real estate. After the fourth quarter of 2008, coinciding with the financial crisis, the percentage of total market value represented by speculative and Ponzi firms in these sectors declines dramatically as they experienced debt deflation because the market revalued these firms’ debt. The one exception is real estate, where this pattern can be seen for Ponzi firms, but not speculative firms. Notably for real estate, hedge finance units never accout for more than 15% of the sector’s market value. There is no particular pattern for utilities or services. Finally, we look at total dollar value for each of the three categories (Fig. 4). Real estate is the first sector to peak in value at the end of 2006, losing value steadily thereafter, primarily from speculative finance units, until July 2008, when the sector lost value precipitously, again mostly from speculative firms, which always account for the majority of the sector’s market value. The next sector to peak, in March 2007, was information, which lost most of its value from hedge finance units. The value of information hedge finance units began to rebound in November 2008, but the value of speculative and Ponzi firms continued to fall, resulting in a continued overall loss in the sector’s market value. Manufacturing, Transportation and Warehousing, and Services all peaked together in July 2007. Hedge firms in Manufacturing and Transportation and Warehousing fell steadily in value thereafter. The value of Services hedge firms continued to grow until March 2008, and fell more sharply than the other two sectors between July and November 2008. Agriculture, Mining and Utilities all peaked last, in April 2008. Most of the loss of value in mining came out of hedge finance units, while for utilities most of the loss came from the speculative finance units which dominate that sector. Agriculture is difficult to characterize because of the seasonal patterns in finance and leverage. The inverted-u pattern is particularly striking for all sectors, but is most dramatic for agriculture.

4. The financial instability hypothesis and the Austrian Business Cycle In spite of the FIH’s Keynesian origins, the FIH and Austrian Business Cycle (ABC) theory are not necessarily mutually exclusive. Credit expansion, hypothesized by the Austrians as the cause of the unsustainable expansion which precedes and lays the groundwork for a recession (Mises, 1912, 1949; Hayek, 1931, 1933, 1935, 1939, 1941), may contribute to the progressive abandonment of prudent borrowing and lending practices and the overleveraging of the financial sector described by the FIH. The main, and perhaps the only, difference between ABC theory and the FIH is that Minsky sees the acceptance of greater risk, leverage, and lower returns as an endogenous byproduct of prosperity, which to the Austrians, can only result from exogenous credit expansion. According to ABC theory, the lower market interest rate brought about by credit expansion systematically biases household consumption-saving decisions in favor of immediate consumption.

R.F. Mulligan / The Quarterly Review of Economics and Finance 53 (2013) 450–459

453

Fig. 1. FIH classification, percent of all listed firms by NAICS sector. Notes: Dark gray = Ponzi firms (top), light gray = speculative (middle), medium gray = hedge finance units (bottom).

454

R.F. Mulligan / The Quarterly Review of Economics and Finance 53 (2013) 450–459

Table 1 T-tests of the FIH 2002–2009. Intercept

Time

Time sq

All

Coefficients Standard error t Stat P-value

75.3470 0.9424 79.9556 0.0000

−1.5734 0.1401 −11.2282 0.0000

0.0484 0.0044 11.0390 0.0000

S&P 500

Coefficients Standard error t Stat P-value

63.6404 1.6782 37.9223 0.0000

−3.5373 0.2495 −14.1748 0.0000

0.1051 0.0078 13.4507 0.0000

Coefficients Standard error t Stat P-value

64.6325 5.5138 11.7219 0.0000

1.3279 0.8199 1.6196 0.1169

−0.0502 0.0257 −1.9561 0.0609

Coefficients Standard error t Stat P-value

81.4321 3.2637 24.9510 0.0000

−2.6604 0.4853 −5.4817 0.0000

0.0802 0.0152 5.2808 0.0000

Coefficients Standard error t Stat P-value

74.7486 1.5316 48.8051 0.0000

−1.7525 0.2277 −7.6950 0.0000

0.0537 0.0071 7.5383 0.0000

Utilities

Coefficients Standard error t Stat P-value

85.3568 3.8089 22.4098 0.0000

0.3849 0.5664 0.6796 0.5025

−0.0213 0.0177 −1.2008 0.2402

Transportation & warehousing

Coefficients Standard error t Stat P-value

76.6115 2.4470 31.3084 0.0000

−2.4233 0.3639 −6.6598 0.0000

0.0760 0.0114 6.6726 0.0000

Coefficients Standard error t Stat P-value

83.3220 1.0406 80.0705 0.0000

−1.0370 0.1547 −6.7015 0.0000

0.0293 0.0048 6.0434 0.0000

Coefficients Standard error t Stat P-value

86.7026 0.9665 89.7037 0.0000

−0.2480 0.1437 −1.7253 0.0959

0.0090 0.0045 2.0093 0.0546

Coefficients Standard error t Stat P-value

69.1555 1.8239 37.9163 0.0000

−0.6479 0.2712 −2.3889 0.0241

0.0158 0.0085 1.8619 0.0735

Agriculture

Mining

Manufacture

Information technology

Real estate

Prof Sci Tech Svs

Sig (t)

***

***

*

***

***

***

***

*

*

Note: The left-hand-side variable in each regression is the percentage of listed speculative and Ponzi firms in each industrial sector. The t-statistic on the coefficient of the square of the time index must be significantly positive to indicate that this percentage fell coming out of the 2001 recession and then rose approaching the 2007–2009 recession. *** 1% significance levels. ** 5% significance levels. * 10% significance levels.

Malinvestment occurs whenever monetary expansion induces firms to systematically overinvest in new production plans which are lower-yielding due to the lower interest rate. Conversely, in terms of the FIH, inflation creates malinvestment by converting hedge finance units into speculative finance units, and speculative finance units into Ponzi finance units, while simultaneously making financial intermediaries able and willing to lend in larger amounts and at lower interest rates. Inflation makes more leveraged firms seem more profitable, and thus better credit risks, and also increases appraisals of the value of their collateral, especially during a period of apparent prosperity. The lower interest rate brought about by the inflationary environment rewards saving less, and given the unchanged rate of time preference, the lower the interest rate falls below the original, pre-credit-expansion rate expressing people’s time preference, the less consumers save and the more they spend on consumption, aka “overconsumption,” an essential element of ABC theory (Mises, 1949, pp. 556 and 564; Garrison, 2001, pp. 68–73). Although

interest rate reductions have both income and substitution effects on consumption and saving, it is clear that the substitution effect dominates. The income effect of a fall in interest rates results in consumers borrowing to finance more consumption spending. By construction, they cannot simultaneously save more. Thus the substitution effect, where the lower interest rate results in a substitution of consumption for saving, must dominate. ABC theory focuses exclusively on growth of monetary assets, and in the context of expanding the money supply, without the exogenous injection of additional credit, there is no way to increase investment spending without increasing saving, thus there is no way to simultaneously increase both investment and consumption. Put another way, credit expansion provides the additional funds required for the simultaneous—though unsustainable—increases in both investment and consumption. Because credit expansion lowers the interest rate, it encourages firms to borrow more as it makes additional funds available to lend, and makes that borrowing cheaper, even in the absence of additional saving. The

R.F. Mulligan / The Quarterly Review of Economics and Finance 53 (2013) 450–459

455

Fig. 2. FIH classification by interest coverage, gross count of firms in each NAICS sector. Notes: Dark gray = Ponzi firms (top), light gray = speculative (middle), medium gray = hedge finance units (bottom).

456

R.F. Mulligan / The Quarterly Review of Economics and Finance 53 (2013) 450–459

Fig. 3. Percent of market value in each FIH category, by NAICS sector. Notes: dark gray = Ponzi firms (top), light gray = speculative (middle), medium gray = hedge finance units (bottom).

R.F. Mulligan / The Quarterly Review of Economics and Finance 53 (2013) 450–459

457

Fig. 4. Total dollar market value by NAICS sector ($ million). Notes: dark gray = Ponzi firms (top), light gray = speculative (middle), medium gray = hedge finance units (bottom).

458

R.F. Mulligan / The Quarterly Review of Economics and Finance 53 (2013) 450–459

lower interest rate also biases consumers’ consumption-saving decisions toward consumption, as it biases producers’ decisions toward investment. An unsustainable boom results because both consumption and investment rise simultaneously, and does so as the amount of saving available to finance consumption and investment decreases. According to ABC theory, if monetary inflation results in increased investment spending, the higher equilibrium quantity of gross investment and lower expected rate of return moves the economy down the investment demand curve where factor productivity falls as investment spending rises – during the inflationary expansion, investment, consumption, employment, and output rise, but real factor productivity falls. Furthermore, the relative ease of obtaining loanable funds swollen by the increase in monetary assets, engenders an unrealistic optimism on the part of entrepreneurial planners (Carilli and Dempster, 2001; Calandro, 2004, 2006). Minsky’s FIH describes a similar process, where optimism leads more and more firms to overleverage themselves to the point where the economy is increasingly fragile and increasingly dominated by speculative and Ponzi finance units. Risk-adjusted yields fall as lending and borrowing expand. Once unrealistic entrepreneurial plans committed to during the expansion phase start proving impossible to complete and coordinate with the plans of others, entrepreneurial planners, finally stung by their overoptimism, become overpessimistic, shifting the investment demand curve to the left. Minsky’s description of debt deflation seems to better capture the onset of recession and financial crisis than the more general description offered by Mises and Hayek. At this point in both ABC theory and the FIH, installed capital is reallocated to next-best uses, and this reallocation process is necessary to liquidate the malinvestment. However, the FIH is broader and more encompassing in one important respect. The unsustainable production structure of ABC theory cannot come about without the exogenous injection of additional money. The FIH goes beyond ABC theory by considering the endogenous increase in demand for all liquid assets, particularly commercial paper and newer, more innovative debt instruments used to finance business expansion outside the banking system. Although in the expansion preceding the 2007–2009 recession, the monetary aggregates grew along with innovative liquidity sources, the distinction is hardly academic. The apparently close regulation of the financial industry may not be sufficient to control the money supply or avoid future recessions. The constellation of malinvestment, overconsumption, and forced saving, with bloated early and late stages of production and desiccated intermediate stages, eventually becomes obvious to entrepreneurial planners, not as an episode of blessed prosperity, but an era of deranged and unfulfillable expectations (Hülsmann, 2001), dominated by Minsky’s Ponzi finance. As entrepreneurs realize their production plans cannot be brought into mutual coordination, and thus brought to completion, the exaggerated over-optimism of the boom transforms into an equally exaggerated over-pessimism of the bust. In terms of the FIH, as speculative and Ponzi finance units impose greater risk on financial intermediaries, it becomes increasingly more difficult for entrepreneurial planners to acquire the financial intermediation necessary to coordinate the production structure. Eventually the external risk—endogenous to Minsky, but to Mises and Hayek the product of exogenous credit expansion—overwhelms entrepreneurial planners’ ability to coordinate production. The unsustainable production structure collapses, lowering output and employment. When the economy reaches the stage of debt deflation/malinvestment liquidation, entrepreneurs face the cost of discarding old installed physical

capital, human capital, and goods-in-process embodied in the old production structure. Davidson’s (2008) evaluation of the current financial crisis is that it has not unfolded according to the FIH. He sees it as originating in the persistent and chronic dependency of mortgage underwriters on the securitization of traditionally illiquid assets, and their inability to maintain that liquidity. However, the unsuccessful selloff of securitized assets and derivatives which occurred in November 2008, was clearly driven by the need of the sellers to raise cash. The FIH was initially framed in terms of conventional bank lending leading to overleverage. The fact that exotic new financial instruments have proliferated to facilitate the same process, seems no real criticism of the FIH—instead it makes Minsky appear nearly prescient. It is also difficult to argue that the financial and housing sectors could have attained any sizable speculative bubble in the absence of credit expansion, and the emergence of various tradable derivatives and securities which allowed for massive growth in these sectors seems to have both depended on and fueled the progressive over-optimism the FIH describes. 5. Conclusion It is very clear that North American stock markets developed between the last and present recessions exactly as Minsky describes in the financial instability hypothesis. Very strong evidence has been presented to support his thesis and world view. Nevertheless, substantial debate needs to occur between Austrians and post-Keynesians over whether business cycles are the endogenous consequence of prosperity, as suggested by Minsky, or arise solely due to exogenous, policy-induced credit expansion. The policy implications are profound. Consider the implications of inflation for the FIH. The stable economy is dominated by hedge-financed firms, those whose ordinary operating cash flows are sufficient to service both debt and principal. In Minsky’s view, the longer the economy operates in a stable regime of hedge finance, the greater the likelihood of endogenous increases in business optimism, leading eventually to the next stage, that of speculative finance, where firms take on debt in anticipation of future growth. If firms are able to continue for any substantial period as speculative finance units, increasing numbers of hedge finance units will imitate them, shifting the balance of the economy as entrepreneurial planners forget the value of conservative safe margining, and are endogenously seduced by the irresistible lure of greater leverage. Finally, firms begin to further overleverage themselves and operate as Ponzi finance units, unable to service either debt or principal out of current cash flow. These firms must sell off assets, or continually seek ever larger rollover loans. The cycle ends in crisis because eventually Ponzi finance units have to sell off their assets, resulting in oversupply and a collapse of asset value. In Minsky’s view, each of the three stages leads endogenously to the next, but consider the impact of inflation, which rewards borrowers at the expense of lenders. To the extent that the progression from hedge finance to speculative finance to Ponzi finance occurs endogenously, this process must be accelerated and exacerbated by inflation. And if Minsky’s FIH would not actually occur endogenously without inflation to drive it, the expansionary monetary policy which drives the ABC would tend to promote such a process. References Binner, J. M., Tino, P., Tepper, J., Anderson, R., Jones, B., & Kendall, G. (2010). Does money matter in inflation forecasting? Physica A: Statistical Mechanics and its Applications, 389(21), 4793–4808.

R.F. Mulligan / The Quarterly Review of Economics and Finance 53 (2013) 450–459 Calandro, J. T. (2004). Reflexivity business cycles and the new economy. Quarterly Journal of Austrian Economics, 7(3), 45–69. Calandro, J. T. (2006). A perspective on post new economy business cycle behavior working paper IBM Global Business Services & the University of Connecticut. Carilli, A. M., & Dempster, G. M. (2001). Expectations in Austrian Business Cycle Theory: An application of the prisoner’s dilemma. Review of Austrian Economics, 14(4), 319–330. Chiarella, C., & Di Guilmi, C. (2011). The financial instability hypothesis: A stochastic microfoundation framework. Journal of Economic Dynamics & Control, 35, 1151–1171. Davidson, P. (2008). Is the current financial distress caused by the subprime mortgage crisis a minsky moment? Or is it the result of attempting to securitize illiquid non-commercial mortgage loans? Journal of Post-Keynesian Economics, 30(4), 669–676. Dos Santos, C. H. (2005). A stock-flow consistent general framework for formal Minskyan analyses of closed economies. Journal of Post-Keynesian Economics, 27(4), 11–35. Dos Santos, C. H., & Zezza, G. (2008). A simplified, ‘benchmark’ stock-flow consistent post-Keynesian growth model. Metroeconomica, 59(3), 441–478. Garrison, R. W. (2001). Time and money: The macroeconomics of capital structure. London & New York: Routledge. Hayek, F. A. (1931). Prices and production (1st ed.). London: Routledge. Hayek, F. A. (1966 [1933]). Monetary theory and the trade cycle. New York: Augustus M. Kelley. Hayek, F. A. (1967 [1935]). Prices and production (2nd ed.). New York: Augustus M. Kelley. Hayek, F. A. (1969 [1939]). Profits, interest, and investment and other essays on the theory of industrial fluctuations. New York: Augustus M. Kelley. Hayek, F. A. (1941). The pure theory of capital. Chicago: University of Chicago Press.

459

Hülsmann, J. G. (2001). Garrisonian macroeconomics. Quarterly Journal of Austrian Economics, 4(3), 33–42. Keen, S. (1995). Finance and economic breakdown: Modeling Minsky’s financial instability hypothesis. Journal of Post-Keynesian Economics, 17(4), 607–635. Keen, S. (2013). A monetary Minsky model of the great moderation and the great recession. Journal of Economic Behavior & Organization, 86, 221–235. Minsky, H. P. (1975). John Maynard Keynes. New York: McGraw-Hill. Minsky, H. P. (1982). The financial-instability hypothesis: Capitalist processes and the behaviour of the economy. In C. P. Kindleberger, & J. P. Laffargue (Eds.), Financial crises (pp. 13–39). Cambridge: Cambridge University Press. Minsky, H. P. (1986). Stabilizing an unstable economy. New Haven: Yale University Press. Minsky, H. P. (1992). The Financial Instability Hypothesis. In Jerome Levy economics institute working paper no. 74 http://dx.doi.org/10.2139/ssrn.161024. Available at SSRN:. http://ssrn.com/abstract=161024 von Mises, L. H. E. (1980 [1912]). The Theory of Money and Credit Theorie des Geldes und der Umlaufsmittel. Indianapolis: Liberty Classics. von Mises, L. H. E. (1998 [1949]). Human action. Auburn: Ludwig von Mises Institute. Moghaddam, M., & Jenson, J. E. (2008). On the death of the resurrected short-run Phillips curve: A further investigation. Cato Journal, 28(1), 139–145. Mulligan, R. F. (2011). An Austrian rehabilitation of the Phillips curve. Cato Journal, 31(1), 87–98. Niskanen, W. A. (2002). On the death of the Phillips curve. Cato Journal, 22(2), 193–198. Reichel, R. (2004). On the death of the Phillips curve: Further evidence. Cato Journal, 24(3), 341–348. Sethi, R. (1992). Dynamics of learning and the financial instability hypothesis. Journal of Economics/Zeitschrift für Nationalökonomie, 56(1), 39–70. Silipo, D. B. (2011). It happened again: A Minskian analysis of the subprime loan crisis. Journal of Economics and Business, 63, 441–455.