INTRODUCTION
Separating signals from noise is a problem in many fields of endeavor. For example, geologists have long been fascinated by the cyclical nature of some sedimentary sequences, like the one pictured above. The cycles visible in that particular sequence in Oregon have been shown to correlate with the Milankovitch (orbital) cycles predicted by theory (David Horn, 2017). Milankovitch cycles are complex, overlapping fluctuations in solar radiation caused by the orbital dynamics of the Earth. There are separate periodicities for the changing tilt (obliquity) of the Earth’s axis (~41,000 years), the eccentricity of the Earth’s orbit around the Sun (~100,000 years), and the wobble (precession) of the Earth (~21,000 years) on its axis (Chart 1). The theory supports detailed hindcasts of Milankovitch cycles stretching all the way back to the Pennsylvanian (aka Carboniferous), some 325 million years ago, and these have been confirmed by detailed field data from many locations and time periods. There are occasional instances (within certain well-preserved rock sequences) of these cycles dating back as much as 700 million years (Precambrian).
Chart 1: The Milankovitch Astronomical Cycles
The problem has always been interpretation, however, because the interaction of geological processes with astronomical and climate cycles has produced complex signals in the rock record with overlapping features and considerable noise (Chart 2). This problem has generally been solved in recent years with the use of bandpass filters which help researchers (e.g., Stephen R. Meyers et al., 2008) separate the main signals from the noise (Chart 3). The treatment is purely mathematical and although sophisticated, yields easily understood results with implications for climate cycles throughout geologic history. The complex pattern of glacials and interglacials during the Pleistocene continental glaciation (over the last 2.6 million years) was probably driven by orbital cycles, which involved fluctuating solar radiation over time; this regulated the extent of the glaciations but did not cause them. The actual prime mover of the entire Pleistocene planetary glaciation event was likely a set of completely different forcing factors, such as the relatively young uplift of both the Himalayas and the Tibetan Plateau, and the reorganization of the world ocean’s circulation system.
Chart 2: Complex Geology, Climatic Cycles, and Overlapping Orbital
Periodicities Often Make Discrimination between Signals and Noise in Milankovitch Cycles Difficult
Chart 3: Bandpass Filters Allow the Milankovitch Signal to Be Separated from the Noise in the Stratigraphic Record
By analogy, in finance and economics we face an even more difficult task in interpreting events and forcing functions. There is great complexity in the signal and almost overwhelming noise obscuring it, much of the time; at turning points this is especially difficult to sort out. Lagging indicators tend to stay positive right into a recession, but are later revised downwards; this makes real time analysis using GDP, employment, and inflation data quite difficult, if not impossible. However, both leading indicators (e.g., housing, the yield curve, etc.) and balance sheet analytics (cf. Dirk J. Bezemer [2010]) can give early warnings of impending recessions or financial crises. Bezemer favored accounting (flow-of-funds or balance sheet-driven) models for examining the causes of the last crisis, and for indications of the next. Bezemer further suggested that mainstream economics failed in 2008 because it has had a blind spot with respect to financial instability. He also suggested that Carmen Reinhart and Kenneth Rogoff (2009; This Time Is Different: Eight Centuries of Financial Folly, Princeton University Press, Princeton, NJ, 491 pp.) got it right when they observed that financial deregulation is one of the best predictors of financial crisis.
CHOOSING ECONOMIC BANDPASS FILTERS
There have been many mathematical treatments of the complex interactions within the economy, especially in recent decades with the advent of Dynamic Stochastic General Equilibrium (“DSGE”) models (Wikipedia, 2018). However, these types of models have severe limitations and have completely failed at major turning points, such as the Great Financial Crisis (cf. Kevin Wilson, 2018a). These failures are in part because of unrealistic model assumptions, and in part because the models capture more noise than signal. Yet, in spite of all the economic noise being generated at various times, there are clear signals discernible at turning points, including the present one, if the correct approach is used to analyze the complex data from the global economy.
Alternative methods using balance sheet-driven heterodox econometric models have actually successfully identified major economic turning points in the past both accurately and well in advance, including the one associated with the Great Financial Crisis (cf. Kevin Wilson, 2018b). These heterodox models have been developed by a number of independent researchers (e.g. Raghuram G. Rajan [2005]; Dean Baker [2006]; Wynne Godley and his co-author Gennaro Zezza [2006]; Michael Hudson [2006]; and William R. White [2006]) who have not generally been taken very seriously by the more fashionable (hence, by definition incompetent) economists of the so-called economics “mainstream” at the Fed and in academia.
However, recently some prominent mainstream economists have awakened to the problems with “DSGE” models and the need for more heterodox solutions. For example, Nobel Laureate Paul Krugman and a co-author published an attempt at a New Keynesian-style model for debt-driven deflationary busts a couple of years after the Great Financial Crisis (Gauti B. Eggertsson & Paul Krugman, 2010). But while Krugman gave credence to the extremely useful ideas of Irving Fisher, Hyman Minsky, and Richard Koo, he also continued to ignore the recent (successful) balance sheet-driven modeling efforts of Baker, Godley, Hudson, Keen, Rajan, White, and Zezza (cf. John T. Harvey, 2018). Brand-new Nobel Laureate Paul Romer has taken a somewhat firmer stand, calling “DSGE” models and their assumptions “post-real” (Paul Romer, 2016). He also refers to modern macroeconomics as having regressed into “pseudoscience” for many years. However, he unfortunately mimics Krugman by avoiding the citation of the successful results of the heterodox modelers. But in partial mitigation of these omissions, Romer did name names of those who fostered the “post-real” approach to macroeconomics: Frank Smets, Rafael Wouters, Robert Lucas, and Thomas Sargent, amongst many others.
The Current Signal from Heterodox Models
Recently I have updated my review of balance sheet-driven heterodox modeling papers that I first wrote about in a three paper series a year ago (cf. Kevin Wilson, 2018c). It would appear that these balance sheet-driven models now foretell a very dangerous period for the global economy in coming quarters. And unlike the experience before the Great Financial Crisis, we now hear warnings from a variety of mainstream economic institutions, including the “BOE,” the “IMF,” the “OECD,” and the “BIS” (Geoff Tily, 2018a; Geoff Tily, 2018b). They all now at least partly recognize that their standard model forecasts prior to 2008 were manifestly wrong, and that the error lies in the construction of the models themselves. The Bank of England (“BOE”) has recently evaluated global vulnerabilities using a more realistic approach than in the past; they’ve identified serious risks in Argentina, Turkey, China, Italy, the US corporate sector, and the European banking sector (Bank of England, 2018). The “BOE” don’t believe a disorderly “Brexit” will cause a financial crisis based on fairly rigorous bank stress tests, but they are concerned about global leveraged lending risks.
The Organization for Economic Cooperation and Development (“OECD”) has written recently (Laurence Boone, 2018) that global growth has peaked, trade is slowing significantly, productivity gains are quite weak (Chart 4), some currencies have devalued significantly (Chart 5), financial risks have become elevated again, and the real economy (especially in Europe) hasn’t regained all of the ground lost during the Great Financial Crisis (Chart 6). The “OECD” also put out another report (“OECD,” 2018) that emphasized the risks to the global economy arising from the recent trade war (cf. Chart 7) and the simultaneous global tightening of central bank monetary policies.
Chart 4: US Labor Productivity Growth Is Weak
Chart 5: EM Currencies Sliding in 2018
Chart 6: Weak Performance of European Economy since the “GFC”
Chart 7: Projected Impact of Trade War on Global Economy
The Bank for International Settlements (“BIS”) has recently reported that the new Basel III banking rules are now finalized and almost fully implemented in most large banks, but they still have concerns about unregulated non-bank lenders and the gaming of the new regulations by the largest banks, which deploy a “window-dressing” approach to quarterly and annual reports (“BIS,” 2018). The “BIS” is also concerned about rising concentration in the asset management industry and rising exposure to valuation losses. Of special concern to the “BIS” is the huge volume of assets held by illiquid ETFs. The “BIS” is also very concerned about the huge potential risk from the derivatives clearing houses or “central counter-parties,” which currently handle $540 trillion in nominal transactions and are themselves vulnerable to failure in a liquidity crisis (Ambrose Evans-Pritchard, 2018).
It was pointed out several years ago that the new banking regulations might not prove sufficient to prevent another crisis (William White, 2014). White attributes this risk to the continued very important role played by the “shadow banking system” and its continued escape from meaningful regulation. White also questions the utility of audits as currently practiced (i.e., practices which ignore the 14,400 off-balance sheet subsidiaries of the top six US banks and the huge number of subsidiaries also held by the European banks). White has further concerns about the differences still extant in international accounting standards, and the extremely unreliable internal risk models used by big banks. White has recently suggested that another currency-mismatch crisis for dollar-denominated emerging market debt could be in store (William White, 2018).
Heterodox economist Dean Baker pooh-poohs the notion that another global financial crisis could happen in the near future (Dean Baker, 2018), but he does accept that the Fed could easily trigger a recession in the current environment. Heterodox economist Steve Keen has recently noted that soaring global debt poses significant risks to the financial system (Steve Keen, 2017). He believes that a crisis is unlikely for the US and the UK, but he expects some trouble in those countries which dodged major crises in 2008 by promoting private debt bubbles (e.g., China, South Korea, Australia, Canada, and Belgium). Keen also laments the continued dominance of nonsensical “DSGE” models in macroeconomics, including at the major central banks. Keen recently suggested that the threat to watch now (Chart 8) is the soaring level of private debt/GDP, which is again above the level of 1929, but still a bit below that of 2007 (Steve Keen, interviewed by Bloomberg, 2018).
Chart 8: Steve Keen’s Diagram of Private Debt vs. Change in Credit
Still, major criticisms of the “DSGE” approach have finally been acknowledged as useful (as already mentioned) and are now driving at least a limited revision to the types of models being used in macroeconomics. For example, some fairly heterodox models now look at liquidity in depth, with the result that systemic risks caused by combinations of high leverage and low liquidity are now capable of being recognized in advance (Raghuram Rajan, 2018). Rajan currently points to high levels of “covenant-lite” leveraged loans as a major source of risk to the system, especially as liquidity is being withdrawn by central banks (Chart 9).
Chart 9: $1 Trillion in Leveraged Loans Are 77% “Cov-Lite”
Well-known heterodox economist Ann Pettifor, who accurately forecast the “GFC” years in advance (e.g., Ann Pettifor, 2003) has also called for major changes in the models used, and the decisions made by, central bankers and economists. She suggests that economists must abandon their disproven (“DSGE”) models, a mission which they have yet to fully accomplish (Ann Pettifor, 2017a). Pettifor has more recently maintained (Ann Pettifor, 2017b) that with global debt ballooning to over 325% of GDP (Chart 10), a steeply climbing corporate debt service ratio on over $8 trillion of corporate debt (Chart 11), and profit margins declining, risks to the system are now quite elevated. She also notes that China’s credit growth has reached the same levels (Chart 12) that other countries reached in all previous financial crises occurring since 1980, including Japan’s (peaking in 1994), Thailand’s (peaking in 1998), the US subprime crisis (peaking in 2008), and Spain’s housing crisis (peaking in 2010).
Chart 10: Global Debt/GDP Growth Since 2002
Chart 11: Impact of Rising Rates on Corporate and Household Debt Service Ratios
Chart 12: Private Credit/GDP in China at Same Level as Prior Crises
Heterodox economists Michalis Nikiforos and Gennaro Zezza (2018) have used flow-of-funds models to evaluate the impact of the recent US tax cut, the likely impact of a potential Trump infrastructure program, and the impact on financial stability of the rising level of private debt. With respect to the latter, and assuming a significant stock market draw-down also occurs, GDP growth would potentially drop 1.25%-1.50% or more from a baseline estimate of 1.50% by end-2019, triggering a deleveraging event. GDP growth would be -1.30% in 2020 in this modeling scenario, suggesting that a recession would occur by the end of 2019 and continue right through 2020. The modeled market draw-down is estimated at about -38%. This would represent a typical recession-related draw-down.
CONCLUSIONS
Risks to the economy and the financial system are significant and are rapidly increasing. Heterodox balance sheet-driven (or flow-of-funds) econometric models indicate that these risks could lead to another financial crisis under certain conditions. Those conditions have not yet been met, but we can observe that the setup is already in place for them to occur. Much depends on the two weakest major players right now, China and Europe. Policy appears to be in error now in the latter, and has long been in error in the former. A perpetual game of “kicking the can” has served both economies over the last decade, but it seems doubtful that similar tactics can continue (without great cost) in the quarters ahead, given the enormous debt loads already in place. The only proven ways to remove the threat of these high and rising debt loads involve the pain of either government austerity programs, currency devaluations, hyperinflations, or defaults (Kevin Wilson, 2016).
Given all of the indicators and risks mentioned above, it would seem prudent to adopt a very defensive portfolio posture if one is an investor, as opposed to a trader. In the end, I believe that those who own long-term Treasuries and gold will make a lot of money in the next 12-36 months, and those holding stocks will writhe in pain at their enormous paper losses. Allowing for human nature, most people will sell closer to the bottom than the top. A bull market for the US Treasury bond is historically the norm under these circumstances (Eric Hickman, 2018), and a strong one is likely again (Van Hoisington & Lacy Hunt, 2018). Given the current long-term sell-off from the January market high, the renewed sell-off from the October market high, and the state of certain national economies (e.g., China, Europe, Japan), it makes sense to invest some money in a gold fund like SPDR Gold Shares (GLD), but only as a short-term hedging trade, not a buy-and-hold position. The iShares Gold Trust (IAU) is an alternative ETF that may be safer for those who want to hold it for a somewhat longer period of time. But the safest form of gold in the event of a true financial apocalypse is physical gold.
Also, for those discounting a possible near-term recession and bear market, some liquid alternatives like the Otter Creek Prof. Mngd. Long/Short Portfolio (OTCRX) could be held to protect assets in the event of a much sharper market draw-down associated with deteriorating economic data. Those in a more defensive frame of mind because of the expected eventual market slide should also hold some long Treasuries, in spite of bearish arguments to the contrary, as a stock market crash would be hugely supportive of bond prices: examples include the Wasatch-Hoisington Treasury Fund (WHOSX), and the iShares 20+ Yr. Treasury Bond ETF (TLT).