“Falsehood flies, but truth comes limping after.”
I wrote this book between January and April 2020, at the height of the COVID-19 pandemic. I am glad that I chose to write a framework book and not a forecast one.
My view of the upcoming decade is clear from the first chapter, and COVID-19 has not altered it: I expect the world to continue deglobalizing. I expect laissez-faire to give way to dirigisme. I expect Europe to remain integrated and in fact accelerate the process. And I expect the US and China to remain at each other's throats – albeit constrained from full economic bifurcation by multipolarity.
The COVID-19 pandemic will likely only reinforce and accelerate these themes – particularly the move away from the Washington Consensus and toward the Buenos Aires Consensus, especially in the US and UK. By the time this manuscript hits the shelves, I suspect the move will be obvious. The time to forecast the apex of globalization, the end of laissez-faire, the Buenos Aires Consensus, and the US–China conflict was back in the early 2010s, not 2020.
Yet the COVID-19 crisis teaches a methodological lesson, particularly when it comes to the limitations of the constraint framework. Like many other investors, I initially underestimated the global impact of the virus: I advised clients to hold cash but not to outright short the market in February.
The constraint framework has a blind spot: the power of market participants' collective psychology to run down the clock on constraint-based forecasts. This weakness is the only feature the constraint framework shares with traditional forecasting methods because they are all at the mercy of time – in any analysis, alpha generation depends on whether the forecast event happens too early, too late, or at just the right moment.
In the case of the constraint framework, the link most vulnerable to time is the zeitgeist of the median voter. Recall from Chapter 4 that political constraints – the most powerful and predictive – hinge on the median voter preference. Usually, this measure is large enough to be material: reliable; quantifiable; and – above all – a median, or moderate, representation of the collective. It is large enough that it discounts irrational or extreme views as outliers with little influence on policymaker action.
The predictability of the link weakens when the collective responds irrationally, en masse, to a single issue. Or when a few Twitterati dominate the discussion by “shouting” over the median voter. The outlier becomes the median in those moments of panic. The zeitgeist veers toward the extreme, and suddenly it's a median voter preference gone wild. Under these circumstances, the panic has the power to delay material, constraint-based outcomes. As a result, it can shift forecast time horizons. In this chapter, I examine this limitation of the constraint framework through the lens of the latest cases of mass irrationality: terrorism and the COVID-19 pandemic.
In early 2014, President Barack Obama said the following about the Islamic State: “The analogy we use around here sometimes, and I think is accurate, is if a J.V. team puts on Lakers uniforms, that doesn't make them Kobe Bryant.”1
That same month, the Islamic State seized Falluja, a major city in Iraq's majority-Sunni Anbar Province, and large portions of Ramadi, the capital of the province. By June 2014, the militant group conquered Mosul, Iraq's second-largest city. In mid-October, the Islamic State was fighting the Iraqi army and various Iran-allied Shia militias on the outskirts of Baghdad.
Obama's statement was eerily similar to President Trump's initial treatment of COVID-19. At a February 28 South Carolina rally, Trump referred to the virus as a “hoax” perpetrated by the media and the Democrats to scuttle the economy.2
My next view will probably get me hate mail from 90% of American readers: I kind of agree with Obama and Trump.3 Yes, the Islamic State was capable spreading of terror, and no, the virus was not a hoax. Obviously, the statements were objectively wrong. However, both Obama and Trump were correct in their initial assessments that materially, the threats were overblown.
They were only wrong because in early 2014 and early 2020, the material reality did not matter. What mattered was the perception of that reality by the public, and thus the markets, for the foreseeable (forecastable) future. The public lost its mind when confronted by the rampaging, decapitating Islamic State and the rampaging, octogenarian-targeting COVID-19 pandemic.4
Obama's flippant comments on the Islamic State cost him, or at least his centrist peers, in the following two years. The Islamic State's rampage across Syria and Iraq and the subsequent 2015 European migration crisis probably gave anti-establishment populists a major boost, contributing to the “Leave” outcome in the UK's 2016 EU membership referendum. It turned Geert Wilders and Marine Le Pen into contenders to lead the Netherlands and France, respectively. It allowed right-wing populist Lega to ascend to power in Italy. While it is difficult to ascertain how much the Islamic State's atrocities contributed to Trump's own electoral success, he certainly benefited from the terror attacks inspired by the group throughout his election campaign in late 2015 and 2016.
All that said, the Islamic State really was a low-quality terror group. While it managed to inspire – and occasionally directly conduct – terror attacks across a swath of the developed world, its methods were of a junior varsity skill level. Compared to Al-Qaeda, the Islamic State's low-quality, high-quantity attacks killed fewer people in the West, failed to target critical infrastructure, and eventually desensitized the Western public. As the public got used to the constant attacks, the “return on investment” of each subsequent attack fell.
On the conventional battlefield, the group proved to be a joke. Yes, the Islamic State had initial success against the demoralized Iraqi army. And it ran amok in the lawless deserts of Syria. But its threat was overstated. Its successes were only possible in the power vacuum of a civil war–ravaged Syria and Iraq. Still, at one point, “people in the know” were seriously contemplating the possibility that the Islamic State could invade Saudi Arabia.
Once the Islamic State was confronted by a loose coalition of the Popular Mobilization Forces (Iran-allied Shia militias in Iraq), Iran's elite Quds force, US special forces and air assets, and Russian air assets in Syria, the group crumbled within months. No nation-state had an interest in the Islamic State's success. Its threat as a terrorist organization was massively overstated because it lacked the kind of command and control that made Al-Qaeda a serious threat.
Writing at BCA Research at the time, I made these forecasts early on in the group's rampage. I bolstered these views with my high-conviction view that the 2015 Europe migration crisis would collapse in a year due to a number of material constraints. It didn't matter. The hate mail poured in. Some clients said I was “nuts” to my face.
Ultimately, the material, constraint-based analysis proved correct. The Islamic State was defeated with minimal effort from the world's powers and local regional players, terrorist attacks inspired by the group subsided (despite the return of many of its fighters home), and the migration crisis dissipated.
Still, I had underestimated the impact that social media would have on collective psychology. The videos of beheadings, migrants streaming across borders, and active shooters running amok across the world's major cities caused voters to believe that the threats of terrorism and mass migration were higher than their material reality.
The unforeseen length of time it took for my forecast to come true demonstrates the constraint framework's potential to be overly forward-looking for its own good. It allows a forecaster to see “around the curve,” the curve being the prevailing narrative in the media. But the markets will respond to the narrative, not necessarily the constraint-based long-term forecast. So, beware of wielding the framework if you have a fiduciary duty to make money.
React as a tactical investor to the narrative and as a strategic investor to the constraints.
This cautionary tale brings me to the COVID-19 pandemic that caused one of the largest bouts of volatility in the history of markets. As of March 2020, I assume it will produce a biblical, but astoundingly brief, recession.
My view of the COVID-19 market sell-off is not going to be popular. I think academics will study it, and the subsequent recession, for centuries as an example of “mass hysteria” on an order of magnitude similar to the one that culminated in the Salem witch trials.
But the investor's job is not to scoff at the irrationality of groupthink from an ivory tower, only to forecast the market. As such, my failure to turn massively bearish at the onset of the crisis in January stings. As my Islamic State forecast also demonstrates, focusing too much on the material reality, as opposed to the zeitgeist of market participants, is a failure of the framework to ascertain when, and for how long, its forecasts will be put on hold.
That said, constraints-based analysis can gauge whether the world will descend into a 1930s-style depression. This call is particularly important here in March 2020, as it is the difference between another 30–50% downside in equity markets – after they fell 36% already – and a bottoming in risk assets on March 23 (the latter being my view).
In the case of the COVID-19 pandemic, there are three broad constraints suggesting that the crisis is ultimately manageable, despite its severity:
Writing in March 2020, this forecast is not obvious. The “flatten-the-curve” narrative has become a dominant approach to fighting the COVID-19 pandemic, which encourages a near-total shutdown of all economic activity. If pursued to its ultimate conclusion, it will cause a depression and thus prove me wrong.
Policymakers across the ideological and geographical spectrum cite the curve-flattening approach as their reason to impose broad lockdown measures. California's Governor Gavin Newsom specifically mentioned the need to “bend the curve” when issuing his mid-March statewide “stay-at-home” order. The view is so prevalent that I've heard it from friends, schoolteachers, and folks in checkout lines at pharmacies. My parents in Switzerland, sister in Milan, aunt in Vancouver … everyone is an amateur epidemiologist and supporter of flattening the curve.
While the World Health Organization (WHO) has most authoritatively pushed for flatten-the-curve policies, the most draconian measures draw inspiration from just two sources. One is an Imperial College London study using available COVID-19 data.5 Another is a blog post.6
The data underlying the Imperial College London study is of poor quality. A study published in the peer-reviewed journal Science estimates that “86% of all infections were undocumented prior to January 23, 2020 travel restrictions” in China.7 Given the paucity of testing in the US and other countries, the spread of COVID-19 is probably monstrously underestimated in America as well, which in turn suggests the US mortality and hospitalization rates are overstated. Another study, published in Nature Medicine, supports my view by suggesting that the mortality rate at the epicenter of the pandemic, Wuhan, was considerably lower than what is being reported by the WHO as of March 2020.8 I suspect that additional studies over the next 12–18 months will confirm that these mortality rates are grossly overstated. My fearless forecast is that the ultimate mortality rate of COVID-19 will prove to be somewhere between 0.1% and 0.3%: higher than the flu, but lower than the widely quoted 2–3%.
The bottom line is that the world is crafting an indefinite public policy based on a linear extrapolation of extremely limited data.9
But not all data is limited. The world does know something about COVID-19. Data plentiful enough to be statistically significant shows that it is ageist. It discriminates by age. The mortality rates across the world – which again, are likely overstated – vary by age cohort. While Figure 8.1 is data from May and June, we had plenty of data out of China and Italy in March to make this conclusion very early (Figure 8.1).
Hospitalization rates are largely in line with the mortality rates, according to the Imperial College London study itself – which again, probably overstates and skews all data to the bearish side (Figure 8.2).
If those above the age of 60 make up 68.2% of symptomatic cases requiring hospitalization, it is prudent to ask why age-specific policies are not part of the arsenal in fighting COVID-19.
Before I pick up on this thread again, let me emphasize my actual area of expertise. I analyze geopolitics and politics from a market and economic perspective. My job is to produce research that projects the macroeconomic implications of political policies and geopolitical events.
So, let me do my job and forecast the implications of a draconian curve-flattening policy.
If G20 economies embrace indiscriminate flatten-the-curve policies, particularly if they do so indefinitely, they will cause a depression. Not a 2008-style Great Recession, but a 1930s-style Great Depression.
Unlike the “Chinese approach,” which was an absolutist model of quarantine and containment, a flatten-the-curve policy accepts that COVID-19 infections will not be brought to a sudden halt. It suggests that countries drag out the pandemic – and presumably social distancing policies – long enough to keep the hospitalizations and tests below the healthcare system's capacity. “Long enough” probably translates to the entirety of the 18 months that it would take to produce a vaccine.
This approach will drag out the economic uncertainty over the course of the summer, perhaps longer. Such uncertainty will end all business investment, which, thanks to the 2019 US–China trade war, was weak to begin with. It will lead to the firing of large numbers of workers in anticipation of lower revenues. No amount of stimulus will deter companies, particularly the small- and medium-sized enterprises that employ the vast majority of American workers, from firing their workforces.
The uncertainty will also lead to drastically lower revenues. If I project the current situation in the restaurant sector, revenues for large parts of the economy will go to zero (Figure 8.3). Such a catastrophe will in turn infect hospitality, airlines, fitness, healthcare (non-COVID-19 related, of course), etc. And as workers in these sectors lose their jobs, they will need fewer goods and services from other sectors. The chain reaction will be of nuclear proportions.
This outlook is a level of economic retrenchment unseen in any recession, not even during the Great Depression. Analysts are dealing with an economic calamity that cannot be reasonably forecast. If data from China is replicated in the rest of the world – and then prolonged due to the open-ended social distancing encouraged by Flattenistas – a depression is almost guaranteed.
A dislocation of this magnitude beyond one or two months would cause a permanent loss of consumer demand. The US consumer – comprising about 15% of global GDP – is particularly at risk. A prolonged recession that mutates into a depression could lead to a “demand hysteresis” (as opposed to merely a labor hysteresis), where consumers permanently retrench due to the combination of the COVID-19 exogenous shock and permanent impairment of their balance sheets. Given that a large swath of America has a negative savings rate, those balance sheets will be impaired within weeks. Days now.
But wait, human lives are in the balance. Wasn't a recession due anyway? So what if stocks fall another 30% and we have to tighten our belts for a decade year or two? It will be worth it if we prevent even one COVID-19 death.
But a deep recession will potentially kill more people than COVID-19, and not just from the well-known recessionary effect of rising suicides. A 2016 Lancet study posited that an excess of 260,000 cancer-related deaths occurred in OECD economies due to the 2008 Great Recession.10 That mortality surplus is just from cancer, and just from the 2008 Great Recession. What would a depression do?
For that, I turn to Greece. A 2018 Lancet study concluded that the Greek mortality rate increased between 2010 and 2016 by 17.8% (Figure 8.4). This was three times higher than that of Western Europe, at a time when worldwide mortality rates were declining.11
A 20% jump in the mortality rate in the US would mean, on average, about 2.5 to 3 million extra deaths annually. And unlike COVID-19, those deaths would not be concentrated among the elderly cohort, as the Greek experience under depression showed.12
And that fatal disaster is only the second-order effect of a depression. For third- and fourth-order effects of a depression, I defer to the politics and geopolitics of the 1930s. As in that decade, in 2020 there is a multipolar distribution of power, an isolationist US, and multiple challenger powers looking to carve out a sphere of influence. With such a similar historical model to go by, it does not take much imagination to extrapolate a 2020s depression to its conclusion: deglobalization, populism, jingoism, and finally a world war.13
Policymakers may therefore pause before handing over the keys of public policy to epidemiologists (and bloggers). Doctors and nurses deserve our support, especially at a time when a pandemic is going to stretch the healthcare system and require healthcare professionals to make Herculean sacrifices. Yet their priorities do not include preventing an economic depression. More importantly, they are not constrained by the possibility of one. They have taken the Hippocratic Oath, which constrains them to focus instead on first-order effects of a medical crisis. Save lives now, let someone else deal with the consequences. They are not trained to worry about – nor do we want them to be distracted by – second-, third-, and fourth-order effects.
But doctors don't make public policy. Policymakers do. And policymakers are constrained by a material reality that includes a depression in its various possible outcomes.
An economic depression is such a massive constraint on policymakers that the trajectory of flatten-the-curve policy will ultimately have to change.
How can policy possibly change when a virus does not? To answer this question, we need to descend into the world of data, which by the summer had given us a fuller picture of the COVID-19 outbreak. As I mentioned above, my March forecast of the eventual mortality rate was somewhere around 0.1% to 0.3%. How could I have confidently made this forecast at the beginning of the outbreak, especially when the World Health Organization (WHO) settled at a much higher number, 3.4%?
My initial forecast was guided by the simple fact that, early on in any outbreak, the number of cases will be massively understated, whereas the deaths will generally not be. In advanced economies, people do not die in the streets. They die in hospitals where their deaths are recorded, and autopsies performed. As such, the numerator of the mortality rate – deaths – was going to be close to reality, whereas the denominator – the total number of infected – would be woefully underreported early on in the crisis.
We also had pretty good data out of China and a trickle of studies that ultimately became a cacophony of research suggesting a much lower mortality rate. As with previous outbreaks – the H1N1 influenza in 2009 and Ebola in 2014 – the initial reporting of the mortality rate was overstated.14
Not only did I fully expect the mortality rate to be massively overstated, my experience tracking the 2014 Ebola outbreak reminded me early in the COVID-19 outbreak to be wary of epidemiological modeling. The Centers for Disease Control and Prevention (CDC) forecast in 2014 concluded that 1.4 million people would get infected with Ebola and that 100,000 would be dead within months.15 Instead, ultimately only 28,646 were infected, with 11,323 deaths. Why the discrepancy?
Obviously, modeling is difficult, particularly in nonlinear environments like a viral outbreak. But what was particularly difficult for the CDC to get right was a change in human behavior. Through excellent public health reaction and individual behavior change, the Ebola outbreak was mitigated. I expected the same would be the case with COVID-19.
But there is another reason why early modeling was so egregiously alarming. I suspect that there was an element of public service in the modeling. They were supposed to be scary so that people would, in the end, change their behavior.
Take the valiant effort by the CDC to convince young people that they are not invulnerable to COVID-19. Just as I was explaining to clients and investors that the median age of death was essentially the same as life expectancy in most countries, the CDC came out with a breathless analysis that the youth was in danger as well. In mid-March, new data from the CDC suggested “young adults” are being hospitalized in the US at higher rates than internationally? As a result, several media outlets – leading with the New York Times – reported that “young adults” are being hospitalized in the US at an elevated rate.16 Figure 8.5 was published accompanying the report.
I could write an entire chapter about this chart! Having excelled during my long career in the “sell side” of the financial industry, I know a manipulated chart when I make see one. The CDC report was either written by a high school student with a poor grasp of mathematics or is an outright piece of propaganda PR.
The data in Figure 8.1 of this chapter, sourced from individual countries' health agencies, segregates hospitalization and mortality rate by age cohorts of 10 years. But the CDC study invents a new category: the “young adult” category of 20–44 years.17 This lopsided category indicates that, according to the CDC, it is empirically acceptable to have one age cohort encompass 25 years where the rest include merely 10.
The 20–44 cohort could have a higher rate of hospitalization for those in the 40–44 age range. It also happens to be the largest demographic cohort, so it would make sense for rates of hospitalizations to be high. But hold on … the chart doesn't even report a “rate.” It reports hospitalizations in absolute numbers. The sheer number of methodological errors in one chart reveal the preference of its creators: manipulation.
There is no other way to explain the CDC reporting than as an effort to get Millennials and Gen Z-ers to take social distancing policies seriously. In this case, the CDC's ends do not justify their means – and the means are not even effective. Publishing manipulated data will not help fight COVID-19, nor will it endear the public to the federal agency that has already bungled America's rollout of testing capacity.
By the summer, another data discrepancy began to appear: a divergence between new cases and deaths. I first noted the divergence in Swedish data, but it appeared to have gone global by June (Figure 8.6).
Now, there may be a lag in deaths, but I doubt that the number would ultimately peak at the same level as early in the crisis. In Sweden, the petri dish that gave us a mirror into the world post-opening, the lag appeared to be almost two months long (Figure 8.7)! Either lingonberries cure COVID-19, the untrustworthy Swedes are manipulating the data, or there is something more structural going on.
If I had to guess, I would suggest that three things are going on:
Ultimately, this discussion is not merely theoretical. As the US opens, the R0 – the rate of infectiousness – is going to rise. Figure 8.9 shows that there is a correlation between the Dallas Fed Mobility and Engagement Index and COVID-19 R0. If hospitalization and mortality rates begin to approximate the elevated levels from April, an R0 meaningfully above 1 will again induce a panic. But if the impact of COVID-19 has been blunted, then the population can slowly desensitize to the risks of a rising R0. In effect, daily new cases become irrelevant from the market's perspective.
As a geopolitical forecaster, I have to bathe myself in nihilist indifference. The novel coronavirus clearly produced a public health crisis that was much worse than the common influenza. However, it was a level of risk, concentrated in a particular age cohort, that the population would eventually become desensitized too. Many investors that I spoke with over the course of 2020 were aghast at this forecast. They themselves did not want to see their elderly parents suffer and die. I didn't either! I love my parents! But that is irrelevant to my job of forecasting the markets.
Not only did data throughout 2020 suggest that the initial forecasts and models of COVID-19 were wrong, many investors also kept focusing on the virus, instead of the policy response initiated to respond to it. This was a major mistake.
Had flatten-the-curve policies from March and April continued indefinitely, as many Flattenistas had hoped, the economy would have been in dire straits. Instead, the outbreak lost potency as hospitalizations and mortality rates apparently began to lag the severity of new infections. Investors who kept trading based on daily new cases missed the point. The gargantuan stimulus raised to respond to the crisis became the most important piece of investment news.
Not only was the economy on pace to recovery, the pace was extraordinary. A V-neck forecast was too cautious. We were experiencing an I-neck recovery. While most investors focused in the early stages of the crisis on the sub-sectors that were “going to zero” – airlines, restaurants, cruise lines, rent-a-car agencies, etc. – the reality is that the combined weight of these sectors in the economy is merely 5.4% of GDP. And even then, they remained at “zero” for only a brief moment, with a recovery of even the most beleaguered sub-sectors underway by May.
Far more relevant than these ancillary parts of the economy where the carnage is indeed vast are durable goods, housing and car sales, which experienced a brisk recovery. Particularly impressive had been the surge in personal consumption on recreational goods & vehicles. It suggested that expenditure not spent on one part of the economy – travel, restaurants, leisure – had been redirected towards another (Figure 8.10).
The recovery in durable goods consumption has been nothing short of extraordinary. While it took durable goods consumption six years to recover to the 2007 highs after the GFC, it took merely two months this time around. Anyone who still thinks that the post-COVID-19 rally had been “retail led” or based on “non-fundamentals” should carefully study Figure 8.11.
What most bears missed is that the main story of 2020 is not COVID-19, but the gargantuan stimulus that has followed it. In particular, that stimulus did not merely rely on the Fed. Monetary stimulus has been subservient to the fiscal, which is real and fundamental. This isn't a “money printing” rally. It is a rally based on fundamentals. The fiscal side of the economy is the definition of economic fundamentals. It is the G in the GDP equation, which states that GDP = C + G + I + NX. There is nothing manufactured, or “printed” about this recovery.
Despite the recovery, I expect $1.5-$3.5 trillion worth of additional fiscal stimulus by the end of the year and likely more stimulus in 2021 to avoid a fiscal cliff. That is how powerful the Buenos Aires Consensus paradigm is in this cycle. Investors should therefore expect new highs in early 2021. Selloffs are opportunities to put money to work and take advantage of the profligate paradigm. This is a fiscal-led market. It is not about the Fed, or COVID-19, valuations, technicals, or anything else. As Snoop Dogg and Dr. Dre would say, it ain't nuthin' but a G thang.
Over the past decade, my clients' number-one criticism of the constraint framework has been that policymakers are not rational. This point is not a major flaw of the framework. Even an irrational policymaker cannot run through a wall. The framework focuses on material constraints, and no amount of irrationality can alter reality.
Had clients instead pointed out that the collective population can be irrational, I would have had a harder time justifying that flaw. As the examples of terrorism and COVID-19 illustrate, fear creates its own reality. There is a reflexive relationship between the median voter – the public at large – and reality. If every voter suddenly demanded a teal Hyundai Sonata, I am pretty sure that they would get one.
A deranged, confused, or simply mistaken policymaker is quickly brought back to reality by constraints – one of which is the median voter. But a hysterical society becomes a material constraint in itself; the median voter reins in policymakers, but society – the median voter – has no such immediate force constraining its actions. As a result, the time it takes for an entire society to return to sanity is unknowable, and impossible to forecast.
If median voters believe that COVID-19 will kill them and their children, that belief is a material constraint. Policymakers have to respond to such a fear with measures that may cause an economic depression. The mass hysteria of voters is an imminent constraint, but an economic depression is further afield, further down the risk curve. As such, even a rational policymaker, grounded in material reality, can use the shorter-term constraint as a reason – and in fact could be forced – to pursue policies that may have disastrous long-term implications.
This chapter tests whether I can successfully use the constraint framework to forecast the behavior of a society. I can.23 In 2014–2015, I held a controversial view that Western populations – particularly Europeans – would become desensitized to terrorism. I believe that a similar desensitization is now occurring with the coronavirus. While COVID-19 is a dangerous business, the mortality rates are almost certainly overstated, and they are highly differentiated by the age cohort. By the end of the summer, I suspect that parents who were worried about their kids' health will be screaming at the district superintendents that children are practically immune.
The real constraint that will break the public's enthusiasm for flatten-the-curve policies is of course not the frustration of homeschooling. The real constraint is money and, in particular, the lack of it. The vast majority of Americans are not savers, despite the aggregate saving rate being elevated. At some point (I suspect with the first round of layoffs), that constraint will kick in and change the zeitgeist of the nation.
In addition to the constraints of childcare and money, the particulars of COVID-19 also represent an important material constraint to prolonged public fear. If this virus were an airborne, Ebola-like virus, my view would not be the same. It is tough to see how the population would become desensitized to a 30% mortality rate. But one that is likely to settle somewhere around 0.3% – three times more potent than the flu – is unlikely to end Western civilization as we know it.
The age of social media has created a high-volatility context for narratives. It is a narrative accelerant, an amplifier of the most extreme ones. In the cases of both the Islamic State and COVID-19, it allowed the alarmists to come to the forefront at the expense of more measured voices. It has also created a viral panic equal to – possibly greater than – the pandemic. But social media narratives die as quickly as they flare up. Their half-life is shorter than media-driven narratives of the past. As such, markets may react with great sensitivity, but they can also recover just as quickly.
In February, as I was writing this manuscript, I felt pretty down on myself. I missed the opportunity to turn mega bearish.
In late March, however, I turned maniacally bullish. Using the framework of material constraints, I posited that data would turn on the virus, given social distancing and non-linearity of human behavior. Meanwhile, my view that the US was transitioning from the Washington to the Buenos Aires Consensus gave me a high conviction view that policymakers would stimulate in a much more meaningful way than in 2009.
When all is said and done, decades from now, investors will look back at 2020 and realize that it was the beginning of a new paradigm. The most important macro chart of 2020 will not be the epi curve of COVID-19, but rather Figure 8.12, a chart that visualizes the transition from the Washington Consensus to the Buenos Aires Consensus. It depicts how, in the months following the 2008-2009 Global Financial Crisis, monetary policy stimulus was quickly followed by the guardrails of austerity. Then, when President Trump stimulated fiscally from 2017 onwards, the extra government spending was immediately followed by the guardrails of a hawkish monetary policy.
In 2020, monetary policy is a slave to the master of fiscal policy. The two lines are moving in lockstep on Figure 8.12. In my view, this will not ease up in 2021, as the incoming Biden administration would not want to usher in its government with a massive fiscal cliff. And if President Trump wins, it is almost guaranteed – given his first term – that he will not turn towards austerity. As we discussed in previous chapters, this has nothing to do with policymaker preferences, but rather the sentiment among median voters that the tenets of the Washington Consensus – namely fiscal prudence – are no longer relevant.
The Buenos Aires Consensus paradigm will lead to a year or two of orgies in asset markets, with new highs likely to follow in early 2021. However, it can't all just be milk and honey from here on out. I would expect inflation to start rising faster than investors expect. While not negative for equities in its early stages, the 2020s will eventually become stagflationary.