War and Peace and War

Inequality, Instability and The Cycle of Human Conflict

Press Play. Read.

Is Communism just a bad idea or is it actually some kind of thermodynamic inevitability? Are stock market implosions that result from the speculative outbursts of moral hazard among Crony Capitalists preventable? Is the rise of Fascism an inexorable consequence of exploding income inequality?

Perhaps it would be more instructive to view the emergence of each of these phenomena as a kind of adhoc signal, an indicator of unsustainable largess on either side of the sociopolitical power spectrum. Perhaps, instead of treating these constructs merely as irredeemable political ideologies — which they are certainly— they might better be understood as the sometimes-inevitable outburst of disequilibrium in a complex system. While history has so far shown that just such disequilibria do eventually decohere into the classical Efficient Market landscape of the General Competitive Equilibrium framework with which we are mainly familiar, when they do persist they generally result in much undue death and destruction.

Without a view to managing the structural-demographic factors which encourage sociopolitical instability, we remain ignorant to the possibility that human agency alone is not the only thing bringing Authoritarianism into existence. And this has a very important implication: mere vocal moralizing won’t solve these kinds of problems; in fact, as Jasney et al. (2015) show in their empirical examination of echo chambers in US climate policy networks this may just lead to their amplification.

The Rise of Fascism as Solar Flare.

Most of the 20th century was a stark lesson in what happens when we stray too far from the classical liberal ideal of individual freedom and small government, but could we have really anticipated the rapid emergence of Authoritarianism? And if we could have then for which particular stretch on the Road to Serfdom was intervention actually feasible?

The problem with the sciences of Complexity is that to truly understand what they entail one must either 1) Disavow oneself entirely of one’s own human agency and transcend fully into Observer mode (much as the nebulous Psychohistorians of Isaac Asimov’s Foundation), or the exact opposite: 2) Become fully immersed with the Zeitgeist of the time and dive head-on into the fiercely quantum game of social life. Both perspectives are replete with their own kinds of biases, but the trouble is that just dabbling once in one or the other necessarily excludes one’s self from switching back between methodologies at a later date.

Anthropologists, for example — Napoleon Chagnon chief among them — have come under incredible scrutiny for portraying indigenous tribes such as the Yamomami in a light that imparts perhaps more about themselves than their subjects. Should Anthropologists get their hands dirty and engage fully with their subjects at a personal level or should they observe carefully from the auspicious silence of a remote laboratory? To be an Observer is to forsake agency and the deep personal connection that entails, but to disengage from neutral Observation is to adopt wholesale the myopia of Solipsism, to see the tree for the forest rather than the forest as an emergent property of many different, interwoven trees.

Despite all that computational social science might imply about the prevalence of a special kind of autism in its proponents, much extraordinary work has been done on the “Observer” side of Social Science lately. In particular, the works of Peter Turchin come immediately to mind. In his attempt to understood history within the framework of a mathematical macrosociology blending the evolution of culture together with economics and a plethora of evidence from carefully maintained historical databases, Turchin has uncovered strong general principles with substantial empirical content. The most impactful of his recent discoveries relates rising income inequality to sociopolitical instability, which, when combined with numerous other demographic parameters too lengthy to discuss here, show how civil unrest can foment into regional conflict and even full-scale civil and total war resulting in the dissolution of entire state societies. [This is a good summary of recent progress in the field of “history as science” (called Cliodynamics) in general.]

To be an Observer is to forsake agency and the deep personal connection that entails, but to disengage from neutral Observation is to adopt wholesale the myopia of Solipsism, to see the tree for the forest rather than the forest as an emergent property of many different, interwoven trees.

I won’t apply Turchin’s work in what follows because it would be remiss to take such fine science and interpret it without so much as an inkling of true rigor, but I do have ideas of my own, threaded together as a consequence of voracious readings from several disparate fields of inquiry, Complexity and Cliodynamics included. I think you’ll come to see that they are at the very least logically consistent and that by following along you will learn somethings very new and quite useful. Besides, understanding Complexity is all about crafting good metaphors, and given that some things simply cannot be predicted — even in principle — I think good metaphors really do suffice.

Waldrop, M. Mitchell. (1992). Complexity: The Emerging Science at the Edge of Order and Chaos. https://www.amazon.com/COMPLEXITY-EMERGING-SCIENCE-ORDER-CHAOS/dp/0671872346

1) Radical New Technology and Industrial Revolution

W. Brian Arthur was the first to truly understand and operationalize what it meant for technology to benefit from increasing returns to scale. Arthur defines “increasing returns to scale” as in general any process “[that]…[once] ahead [becomes] further ahead, for that which loses advantage to lose further advantage.”

Before Arthur, economists dismissed the notion hand-in-fist. The running joke inside traditional Neoclassical economic modeling circles was that “there could never be such as a thing as finding a $100 bill on a sidewalk because somebody would have already picked it up”. Basically, all economic anomalies average out instantaneously, monopolies always dissipate and everything else is subject to the ruthless indifference of entropy and the diminishing returns to scale that inevitable competition implies. By contrast, attaining scale is the quintessential objective of any contemporary Venture Capital-funded platform technology that comes out of Silicon Valley; crushing all competition is the singular imperative that surely drives all powerful technology companies toward the Holy Grail of Singularity. If your product doesn’t exhibit intrinsic virality, you’re not getting funded. End of line;

For more on the subject, Arthur outlines his theory in a very accessible Harvard Business Review article.
John D. Rockefeller (1884)
Good management consists in showing average people how to do the work of superior people. — John D. Rockefeller
Peter Thiel (2014)
“CREATIVE MONOPOLY means new products that benefit everybody and sustainable profits for the creator. Competition means no profits for anybody, no meaningful differentiation, and a struggle for survival.” — Peter Thiel (From Zero To One, 2014)

Both Rockefeller and Thiel advocate a narrative in which those who already have must earn more in order to benefit those who have less (I’ve touched on the fallibility of Hindsight as applied to New Economic Thinking here.) To be fair to Thiel, he does acknowledge that a lack of competition can not last forever, though in the same vein he seems to conclude that insomuch as a monopoly is generating creative new technologies they should be protected: hyperbolic revenues today mean intellectual property developed tomorrow that in the long run benefits everybody.

Such a claim, however, is inherently unscientific. According to Karl Popper, only falsifiable claims should be considered to be truthful. Unfortunately for techno-optimists, monopoly creates a paradox in that current levels of productivity cannot be compared to that which otherwise could have emerged in its absence. In effect, you would need a Time Machine to validate the claim that monopoly product A is superior to every conceivable alternate product that may have otherwise taken-off. And you’d have to make that trip several thousand times or more to compare significant differences between each world line.

By contrast, we can show with a pen, a piece of paper and a simple two-variable statistical simulation that an economy which resembles the standard model of Perfect Market Competition more closely (although definitely not too closely) maximizes both consumer and producer surplus in general (because sufficient competition generates only one or very few equilibria). As a result, the burden of proof falls to Monopolists and Oligopolists to prove that their vision of the future maximizes welfare in much the same way, except for all possible equilibria across all possible timelines. Unfortunately, we do not yet have the computational capacity to determine such a thing as this would require quantum calculations on a magnitude spanning more seconds than will ever take place in the Universe. On the plus-side, an absence of evidence to the contrary is highly convenient for incumbent technocrats. All they have to do is point to how much better their products are than those which have since come to pass. Is Google doing better than Altavista? Of course. Could Altavista have done better? Maybe, but there’s no way to prove that.

One common counter-argument is that accelerated scale also implies accelerated decline. So far, however, the race seems to be more about who’s losing right now rather than if some geeked-out garage band might produce the next big winner.

I will concede to one point: Thiel rightfully believes that society is once again situated at the precipice of a feverish new wave of innovation consisting of radical new technologies, and those technologies (if left unhindered by progressive taxation and regulation) are the key to unlocking immense opportunities for human flourishing.

40 years after Standard Oil was convicted by the federal government of monopolistic practices and divided, Rockefeller had already become America’s first Billionaire. And much as our own Chan Zuckerburg or Bill & Melinda Gates Foundations aim to eradicate all disease and other lofty goals, Rockefeller and his vast wealth would thereafter leave an indelible mark on medicine, art, education and scientific research for another five decades and more. But in his wake, and in the wake of other so-called “Robber Baron” magnates of the late 19th century, is a very prescient story about social disintegration.

2) Income Inequality and Wage Deflation

The Second Industrialization Revolution ended in the final third of the 19th century (late 1800’s) and consisted mainly in the production of steam engines, railroad networks, the mechanization of agriculture, and the coal-powered electrification of manufacturing. The scalability of these technologies is eerily similar to the kind that we see in today’s tech environment, in that both are economically and socially disruptive, and most importantly contribute to rapidly widening income inequality through automation. Many are now calling our own time the Third or Fourth Industrial Revolution (depending on who you ask) but that we are currently experiencing one is not contested.

Cassidy, John. (2014). Piketty’s Inequality Story in Six Charts. The New Yorker: http://www.newyorker.com/news/john-cassidy/pikettys-inequality-story-in-six-charts

So where does above-normal Income Inequality come from? And what’s with the trend-reversal starting in late 1970's? Not only is current U.S. income inequality higher than it was in the run-up to the Great Depression, but on a relative basis, it’s higher even than it was during the late Roman Empire (Berman, 2011). Why is it that the more inequality increases the more wages stagnate? Piketty’s central theorem is that capitalism has a “contradiction”: when the rate of return on capital exceeds the rate of economic growth, inequality tends to rise. So what’s driving recent exorbitant increases in the rate of return on capital? Let’s take a look at a few metaphors:

America’s Agricultural Revolution:

The cities weren’t spared — far from it. As rural incomes fell, farmers had less and less money to buy goods produced in factories. Manufacturers had to lay off workers, which further diminished demand for agricultural produce, driving down prices even more. Before long, this vicious circle affected the entire national economy.
The value of assets (such as homes) often declines when incomes do. Farmers got trapped in their declining sector and in their depressed locales. Diminished income and wealth made migration to the cities more difficult; high urban unemployment made migration less attractive. Throughout the 1930s, in spite of the massive drop in farm income, there was little overall out-migration. Meanwhile, the farmers continued to produce, sometimes working even harder to make up for lower prices. Individually, that made sense; collectively, it didn’t, as any increased output kept forcing prices down — Joseph Stiglitz (Yglesias, 2012), emphasis my own.

Could Stiglitz’ story function just the same by replacing “rural” with suburban “incomes?” “Farmers” with white-collar workers; “goods produced in factories” with services produced by algorithms; “manufacturers” with data conglomerates? I’d say the analogy is strongly worth considering.

Here is where I have to call out people who cry “Luddite!”, particularly those in coastal enclaves such as Silicon Valley (on Market Street) or New York (on Wall St). Are “Luddites”, or suburban lower-income workers really afraid of technology per se? Or are they afraid of the social and economic exclusion that just such a lack of access to new forms of capital spells to their shared fate?

3) Sociopolitical Instability and Economic Depression

If Radical New Technologies lead to Scale Monopolies which lead to Income Inequality which leads to Wage Deflation, Sociopolitical Instability and a Reduction in Aggregate Demand (NBER, 1993), it is precisely the lowest rungs of society that have the most to lose and the most to fear. This is probably the group that you’d want to piss off the least in times of duress. This is also probably the group that will likely elect Authoritarians in the hope that they might relieve themselves of precisely this kind of existential dread. Chew on that for a moment, Lame-stream Media!

Remember how vocal moralizing leads only to echo chambers (Jasney et al., 2015)? Despite how many rural “White” “Nazis” the press feels entitled to denigrate by smearing the label “Deplorable Skinhead” for every click-bait headline they can come up with, these people aren’t going to shut up and take uncompromising social exclusion for an answer. You took away their health, they sat patiently, likely confused more than anything and died (Marmot, 2005). You took away their jobs, they elected Donald Trump. You took away their voice, they started marching. But to take away their dignity, by carelessly brandishing just about any suburban Southerner as “Nazi”… well that’s just about the last leg:

The Truth.

During the previous global “disintegrative” phase following 1929’s Black Tuesday, or The Great Crash, America was fortunate to have elected a Progressive in the form of Franklin D. Roosevelt over incumbent Herbert Hoover, whose party’s penchant for austerity at the time would have probably deepened the Depression (Republican Party Platform, 1932). Instead of Civil War or a Communist Revolution, America worked at fever-pitch to maximize employment in the face of ailing aggregate demand.

Public works programs restored the social dignity of the despondent unemployed. Infrastructure spending opened up rural America to further economic development. These were corrective measures that likely prevented total chaos. But while the stage was set for U.S. economic reform, in part because the government was able recall extensive foreign loans from disenfranchised parts of Europe, nations such as the Weimar Republic in Germany were not so lucky. In effect, American recovery implied further immiseration for Europeans, and Germans in particular because spending for some meant austerity for everyone else:

By 1924, after years of crisis management and attempts at tax and finance reform, the [Weimar] economy was stabilized with the help of foreign, particularly American, loans. A period of relative prosperity prevailed from 1924 to 1929. This relative “golden age” was reflected in the strong support for moderate pro-Weimar political parties in the 1928 elections. However, economic disaster struck with the onset of the world depression in 1929. The American stock market crash and bank failures led to a recall of American loans to Germany. This development added to Germany’s economic hardship.

4) War or What? Solutions.

This is where my own cyclical theory has a chance to veer off into abject conspiracy but I won’t necessarily take that opportunity. I will, however, outline how it could do so on its own. At this stage in the game society has two largely diametric solutions:

  • 1) An aggressive increase in collective redistribution. This can be achieved either through: A) a politically preemptive situation reminiscent of FDR’s Socially Progressive “New Deal” and other Keynesian-inspired national works or spending programs (see Krugman’s End This Depression Now!) that leverage fiscal policy in order to balance national consumption and investment (Bar Yam, 2017); B) a UBI (Universal Basic Income) or Progressive Pigovian tax-break for families and small-to-medium-sized businesses, financed by the top 3% of tech-monopoly revenue; C) the worst-case scenario, via some kind of “Communist” or Proletariat Revolution: the liquidation of the Capitalist class and total seizure of its assets to be redistributed to the working labor share of the population (absolute worst long-run outcome, albeit “short-term rational” given the impending constraints of a worsening sociopolitical environment).
  • 2) A wholesale reduction in the supply of labor. The ethically impermissible (evil) solution would be to manufacture the conditions of war so as to actually re-inflate average wages by killing as many people as physically possible. A nefarious cabal of Deep State actors could conceivably create a war in which the result was a long-lasting integrative phase of social re-cohesion resulting from a population-adjusted higher average wage and the stochastic redistribution of capital that naturally ensues in its aftermath. A war would alleviate the problem of extreme income inequality that results naturally from the invention of Radical New Technologies and the Scale Monopolies that they generate. This solution would also allow Capitalism to continue unhindered, and perhaps even further amplified. Our best estimates for World Wars I and II place relative fatalities at on average about 3-5% of total Earth population. Is this enough to re-create global prosperity? I won’t pretend to even want to know. Such a path is clearly beyond moral reproach despite the apparent ends. But is such a scheme possible? Probably yes, except that “natural” war would likely unfold much the same way, and perhaps the two methods of generating it become indistinguishable as its inevitability becomes more readily apparent (war as both object and consequence of inequality).

5) Bonus: Everybody Loves Wine

In any case, as the rate of return on capital soars beyond the rate of economic growth, wages deflate in proportion to increases in the socioeconomic leverage of the capital class. As capital becomes more powerful, it can employ more workers at a lower price, all the while lobbying for immigration reform that further favors their bottom-line. Therefore, it is quite obvious that immigration in bad times will lead a decline in the value of domestic labor. If times are sufficiently miserable, this will lead to political polarization, such as the kind we are witnessing today in America’s South. Proxies for this kind of behavior can be observed as far back as the Roman Empire (Turchin, 2013).

Apparently, there is a strong positive correlation between wine-drinking and income equality (not exactly obvious, hey?). The idea is that wealth is typically distributed accordingly to a power law (the famous 80/20 fractal) in just about any society ever observed; on the other hand, natural calamity is by and large a stochastic affair: large-scale plagues, viral epidemics, and natural disasters did historically affect any particular income group about as equally as any other (Harper, 2015). When wealthy estate-holders died in these calamities, their wealth was suddenly “up for grabs,” or at least their families had a much harder time defending their savings (and their market power). This, in combination with a general reduction in the labor supply meant higher wages for most people for some time after each disaster.

The population of the Roman Empire grew rapidly during the first two centuries up to 165AD. Then came a series of deadly epidemics, known as the Antonine Plague. In Roman Egypt, for which we have contemporary data thanks to preserved papyri, real wages first fell (when the population increased) and then regained ground (when the population collapsed). We also know that many grain fields were converted to orchards and vineyards following the plagues. The implication is that the standard of life for common people improved — they ate less bread, more fruit, and drank wine. The gap between common people and the elites shrank.
Naturally, the conditions affecting the labour supply were different in the second half of the 20th century in the US. An important new element was globalisation, which allows corporations to move jobs to poorer countries (with that ‘giant sucking sound’, as Ross Perot put it during his 1992 presidential campaign). But none of this alters the fact that an oversupply of labour tends to depress wages for the poorer section of the population. And just as in Roman Egypt, the poor in the US today eat more energy-dense foods — bread, pasta, and potatoes — while the wealthy eat more fruit and drink wine. — Peter Turchin. (2013).

6) A Longish Peace (Integrative Phase)

One common argument against Keynesian economic intervention in the Great Depression was that after the war the U.S. federal government reduced its spending by two thirds, allowing for a dynamic and vibrant market to spring up once more, precipitating what is commonly known as the Post-War Boom.

Except most of the decline in government involvement was due merely to a relative deescalation in existing military spending, and as such other areas grew far faster despite American expansionism in the decades that followed (U.S. Government Publishing Office, 2017). The major misconception of the time was that an arbitrary increase in government spending, for no reason other than accidental political ideology, deepened the Depression. This is about as stupid as the idea that the election of Donald Trump was due to no reason other than the totally acausal, brazen, cavalier and random emergence of racism, xenophobia, bigotry and hatred all across America. This is beyond ridiculous, but mainly intellectually dishonest.

No, American post-war prosperity was more likely a result of its newly established socioeconomic hegemony, which provided disproportionate reach in global markets that could not have otherwise been accessed at the scale in which they were sequestered from underutilized Axis capital after World War II. So as the returns to capital were shared more broadly per American citizen this meant that at least temporarily social integration could begin once more. In moderation, redistribution has been proved beyond equivocation to basically have benign effects in average times, and extremely positive effects during the very bad (Ostry et al., 2015). Redistribution is especially effective in ramping up growth during periods of historically above-average income inequality (Ostry et al., 2015). Furthermore, a climate of social cohesion and lower inequality (due to a larger Middle Class) leads to faster and more punctuated economic innovation (Acemoglu et al., 2012).

It is this same increase in innovation during periods of social stability that feeds back into our cycle, beginning anew with the emergence of Radical New Technologies that lead to Scale Monopolies, Income Inequality and so on. It is precisely at this juncture, when instability is low, innovation is relatively high, and growth is just about right that we as a society should resist the temptation to transcend our present riches, no matter close and how delicious the taste of that impending new Singularity…lest History repeat itself.

Without venturing too much further, I would argue that it is precisely at this juncture, when instability is low, innovation is relatively high, and growth is just about right that we as a society should resist the temptation to transcend our present riches, no matter close and how delicious the taste of that impending new Singularity appears…lest History repeat itself.

5) History as Tao

It was said that History enters when the space of the Possible is vastly larger than the space of the Actual. So it follows that the History that cannot be told will once again come to pass.

Despite Francis Fukuyama’s famous assertion that “History is over”, the notion that we’ve reached a Singularity in terms of political change is deeply misguided, terrifying really. It should come as no surprise that the oft-chanted mantra “this time is different” is what has and always will allow for History to repeat itself. This time, however, it won’t be just millions but billions who would perish.

It doesn’t have to be this way. For the first time in human history, we really do have the tools to help us steer a path into a brighter future. The merging of modern thermodynamics with economic macrosociology via large-scale agent-based simulations, Cliodynamics and other non-linear dynamical sciences is providing a plethora of new evidence in support of the meso-dynamic view of Complex adaptive systems, something that “Saltwater”economists of times past simply could not operationalize at a mathematical level.

John Maynard Keynes, with hindsight perhaps one of the first Complexity theorists, far ahead of his time and exasperated by the status quota of Depression-era economic thinking once stated:

The long run is a misleading guide to current affairs. In the long run we are all dead. Economists set themselves too easy, too useless a task if in tempestuous seasons they can only tell us that when the storm is past the ocean is flat again.

An easier way of saying this follows:

Sometimes, what happens in-between is just as important as what happens in the end.

That “concursus (Tempestuous Seasons) over equilibrium (the ocean flat)” should come to define our time more than any other antiquated metaphor such as the Invisible Hand or “equilibrium” would be a testament to the maturity of our now Global Civilization. And in the absence of predictive power in times of Chaos, we must follow history carefully, placing confidence in our ability to employ the right analogies to their best affect at the best possible time. By applying this new-found knowledge, we may just be able to avoid the unimaginable inevitable for next 10,000 years.

John Maynard Keynes — Keynesian economics as non-equilibrium meso-dynamic view of Complex adaptive systems?

References

Yaneer Bar-Yam, Jean Langlois-Meurinne, Mari Kawakatsu, Rodolfo Garcia, Preliminary steps toward a universal economic dynamics for monetary and fiscal policy, arXiv:1710.06285 (October 10, 2017).

U.S. Government Publishing Office. (2017). 20th Century Defense Spending. http://www.usgovernmentspending.com/spending_chart_1900_2020USp_XXs2li111mcn_30f_20th_Century_Defense_Spending

Facing History and Ourselves. (2016). 1929: A Turning Point During the Weimar Republic. https://www.facinghistory.org/weimar-republic-fragility-democracy/readings/1929-turning-point

Turchin, Peter. (2016). Ages of Discord: A Structural-Demographic Analysis of American History. https://www.amazon.com/Ages-Discord-Peter-Turchin/dp/0996139540

Harper, Kyle. (2015). Pandemics and passages to late antiquity: rethinking the plague of c.249–270 described by Cyprian. Journal of Roman Archaeology 28, 223–260: https://www.cambridge.org/core/journals/journal-of-roman-archaeology/article/div-classtitlepandemics-and-passages-to-late-antiquity-rethinking-the-plague-of-span-classitaliccspan249270-described-by-cypriandiv/BF0A142F1864D05FDC0C3439D182B7B0

Jasney, Lorien; Waggle, Joseph & Dana R. Fisher (2015). An empirical examination of echo chambers in US climate policy networks. Nature Climate Change 5, 782–786: http://www.nature.com/nclimate/journal/v5/n8/full/nclimate2666.html?foxtrotcallback=true

Ostry, Johnathan D., Berg, Andrew, and Charalambos G. Tsangarides. Redistribution, Inequality, and Growth. IMF: https://www.imf.org/external/pubs/ft/sdn/2014/sdn1402.pdf

Cassidy, John. (2014). Piketty’s Inequality Story in Six Charts. The New Yorker: http://www.newyorker.com/news/john-cassidy/pikettys-inequality-story-in-six-charts

Turchin, Peter. (2013). Return of the oppressed: From the Roman Empire to our own Gilded Age, inequality moves in cycles. The future looks like a rough ride. Aeon: https://aeon.co/essays/history-tells-us-where-the-wealth-gap-leads

Acemoglu, Daron; Robinson, James A., and Thierry Verdier. Can’t We All Be More Like Scandinavians? Asymmetric Growth and Institutions in an Interdependent World. NBER: http://www.nber.org/papers/w18441

Yglesias, Matthew. (2012). Joe Stiglitz on the Great Depression. Slate: http://www.slate.com/blogs/moneybox/2011/12/14/joe_stiglitz_on_the_great_depression.html

Burman, Jillian. (2011). U.S. Income Inequality Higher Than Roman Empire’s Levels: Study. HuffPost: http://www.huffingtonpost.com/2011/12/19/us-income-inequality-ancient-rome-levels_n_1158926.html

Marmot, Michael. (2005). The Status Syndrome: How Social Standing Affects Our Health and Longevity. N Engl J Med; 352:1159–1160: http://www.nejm.org/doi/full/10.1056/NEJM200503173521126

Arthur, W. Brian (1996). Increasing Returns and the New World of Business. Harvard Business Review: https://hbr.org/1996/07/increasing-returns-and-the-new-world-of-business

NBER. (1993). Economic Instability and Aggregate Investment. http://www.nber.org/chapters/c11002.pdf

Waldrop, M. Mitchell. (1992). Complexity: The Emerging Science at the Edge of Order and Chaos. https://www.amazon.com/COMPLEXITY-EMERGING-SCIENCE-ORDER-CHAOS/dp/0671872346

Republican Party Platforms. (June 14, 1932). Republican Party Platform of 1932. Online by Gerhard Peters and John T. Woolley, The American Presidency Project: http://www.presidency.ucsb.edu/ws/?pid=29638.