American Political Cycles
The American historian Arthur Schlesinger Sr. (1888-1965) proposed a cyclical concept of American politics in which the political "spirit of the times", or zeitgeist, oscillated between "liberal" and "conservative" eras . Although he never formally proposed a model to explain the cycle, Schlesinger suspected that the dynamics of political organization itself was responsible for the timing. A successful political party or movement takes about 15 years to define its agenda, mobilize its resources, implement its policies as best it can, and obtain the inevitably less-than-hoped-for results . This movement proceeds through several fairly predictable stages: growth and vitality under a charismatic leader, a period of mature leadership, and then a gradual decline as supporters tire of the message. With decline, the baton of leadership passes to the opposition. The result would be alternating periods of ascendancy that should last about 15 years.Figure 1. Conservative political waves
I attempted to track this cycle empirically by assembling a list of political events that I labeled as either liberal or conservative. I then calculated a moving sum of conservative events divided by the sum of all events over the prior 15 years and plotted the result in Figure 1. The 15-year summation period was selected based on the political organizational dynamic explanation. This ratio presumably measures swings in the trends in political zeitgeist over time, with a rising "conservative ratio" implying a conservative era and a falling conservative ratio a liberal era. A crude presentation of this methodology can be found in my second book, The Kondratiev Cycle.  A more extensive treatment is provided in my forthcoming book Cycles in American Politics, expected to be out this fall.
Note that the method I employ measures liberalism and conservatism independently of the political party that was in power when the event occurred. For example, Schlesinger's son Arthur Jr., who continued his father's studies, predicted a McGovern victory in 1972 because the country was in a liberal era at the time and Democrat McGovern was more liberal than Republican Nixon. Nixon won the election. Thus, although Nixon was certainly more conservative than McGovern, his administration was more liberal than that of either President Eisenhower or President Reagan and thus reflected the liberal zeitgeist, despite Nixon's own preferences. Thus, the type of era doesn't actually reflect what party is in power, but rather, what types of policies are pursued.Figure 2. Federal government spending as a percent of GDP 1900-2003
Another indicator for this cycle is government spending. Figure 2 shows Federal government expenditures as a percentage of GDP since 1900. This graph vividly illustrates the concept of liberal government activism. The idea first appeared during the Progressive era, but it reached fruition during the New Deal. Both non-defense and overall government spending rose dramatically during the New Deal liberal era. During the 1946-1961 conservative era, non-defense spending did not rise although overall spending did, reflecting the massive military spending of the early Cold War period. Both overall spending and non-defense spending rose steadily in the 1961-1980 liberal era. In the Reagan conservative era non-defense spending held at roughly constant, while overall Federal spending declined. In recent years both have again been on the rise, providing evidence that we entered a liberal era around 2000.
Having described the political cycle, the next step is to look for relations between this cycle and other cycles of interest I have described in previous articles. Table 1 shows several other cycles compared to the political cycle. Secular bull markets (in italics) and secular bear markets are listed in the column marked stocks. Prior to 1861 entire Stock Cycles are listed rather than the secular trends. Recall that whole Stock Cycles correspond to Kondratiev waves  so the comparison being made before 1861 is between the K-cycle and the other cycles. Kondratiev upwaves are denoted in bold.
Since 1921, conservative eras have been correlated with secular bull markets. With the end of a secular bull market in 2000, a political shift towards liberalism should occur after 2000, if the post-1921 correlation is still valid. Such a shift seems rather counterintuitive with the electoral victories by Republicans in 2000, 2002, and 2004, and by the conservative views of President Bush. Of course, the whole purpose of cycle analysis is to spot developing trends before they become obvious. But before I will explore the idea that we have entered a liberal era with the election of President Bush, further investigation into the validity of the proposed cycle timing is warranted.Table 1. Liberal-conservative cycles in American politics compared to other cycles1
|C||1816-1828||1815-1843||1824-28 (Nat Rep)|
|C||1840-1860||1843-1861||1844-1860||1840-44 (Whig)1844-1860 (Dem)|
|C||1980-2001||1982-2000||1984-2001||1980-1992 (Rep)1992-2000 (Dem)|
3Values in italics represent conditions favorable to conservatives: conservative politics, secular bull market, capital returns rising relative to wages, falling unrest. Bold indicates Kondratiev upwaves.
Table 1 also presents turnings obtained from William Strauss and Neil Howe's generational cycle they call the saeculum [4,5]. The dates I give are slightly different than their dates. They are based on an 18-year average turning length since 1820. The orthodox position is that turnings are 20 years long, but are running 15 years ahead of schedule because the Civil War turning was unusually short. My reasoning for preferring 18-year turnings is given in my new book Cycles in American Politics. For now, I will simply point out that in my view turnings average 18 years in length and are aligned with secular stock market trends, which also are 18 years long.
Since the 1820's, when the modern saeculum emerged, social moment turnings (bold) have been associated with liberal political eras. Assuming that a Crisis turning did begin with the terrorist attack on September 11, 2001, the alignment with the saeculum also supports the idea of a liberal political era beginning in 2001. The problem with using the saeculum to project other cycles is that the saeculum lacks real-time (or close to real time) markers for cycle position. That is, it is difficult to know whether or not we have entered a new turning for quite a long time. Strauss and Howe were able to detect the beginning of the Millennial generation in 1982, nine years after the fact in Generations . They identified the 1984 beginning of the recent unraveling 13 years after the fact in The Fourth Turning . In that same book they projected the start of the next Crisis turning for about 2005, which is still a year away. As early as 2002, the idea that we began a secular bear market in 2000 was not controversial amongst stock market analysts, whereas the idea that a Crisis era began in 2001 is still hotly contested amongst generation cycle enthusiasts today.
One way to deal with this uncertainty is to apply the 18-year standard generational length as a ruler to measure out elapsed time from earlier cycle markers for which there is no disagreement. Laying out four 18-year rulers from the uncontested beginnings of the previous Crisis in 1929 and eight rulers from the beginning of the Civil War Crisis gives estimates of 2001 and 2004 for the start of a new Crisis. Looking at the most recent turnings of each type, if I lay out three rulers from the beginning of the High in 1946, two rulers from the 1964 beginning of the Awakening and one ruler from the 1984 beginning of the Unraveling, I obtain 2000, 2000, and 2002 as estimates for the beginning of a new Crisis turning. Thus, there are five projections, all clustered in the 2000-2004 time frame, with a median and average value of 2001. The appearance of the unsettling terrorist attack in that same year provides an excellent "trigger event" for dating the start of the Crisis turning.
Direct application of the ruler to the political cycle itself also suggests that the post-1980 "Reagan era" of conservative zeitgeist had grown "long in the tooth" by 2001. Assuming the Reagan era has extended beyond 2001, this era is longer than all but one of the seven previous conservative eras. Extending the conservative era beyond 2004 would make the 43+ year length of the most recent political cycle longer than all but one of the 12 previous political cycles. If the Reagan era continues, it will become the longest cycle in history in 2008.
Eras of political dominance by one party or the other are also presented in Table 1. Conservative party eras are in italics. Party power is not correlated with the political cycle.
What do cycles say about the 2004 election?
Table 2 shows electoral results as a function of political eras since 1856. In the House, Democrats gain seats 59% of the time and lose them 41% of the time during liberal eras. During conservative eras the corresponding figures are 44% and 54%. In the Senate Democrats gain seats 50% of the time and lose them 38% of the time during liberal eras. During conservative eras the corresponding figures are 49% and 41%. In Presidential elections, Republicans are strongly favored (71%) during conservative eras whereas Democrats are only slightly preferred (56%) during liberal eras.
This would suggest a slightly better than even chance for Democrat Kerry to win in the 2004 election. But Republican candidate Bush is an incumbent. If I focus on just Republican incumbents during liberal eras since 1856 I find that they have won three times and lost twice. This suggests a 60% probability that Bush would win re-election if today is a liberal era. If today is a conservative era, a Bush victory would also predicted. This shows that a Bush victory would give no confirmation of what sort of era we are in. On the other hand, a Kerry victory is only 29% likely (Table 2) if the Reagan conservative era is still in operation. Incumbency has had no effect, Democrats have beat incumbent Republicans two out of seven attempts (29%) in conservative eras. So if Kerry had won, it would have provided some support for the idea that a liberal era had indeed begun.Table 2 Electoral results since 1856 as a function of era
|House||Conservative Era||Liberal Era|
|Democrat gain (avg # of seats)||18 (35)||19 (30)|
|Democrat loss (avg # of seats)||22 (29)||13 (34)|
|Senate||Conservative Era||Liberal Era|
|Democrat gain (avg # of seats)||20 (4.8)||16 (5.4)|
|Democrat loss (avg # of seats)||17 (5.9)||12 (5.9)|
|Presidency||Conservative Era||Liberal Era|
Congressional elections might provide some insight. Table 2 shows that Democrats tend to pick up seats in both the House and Senate during liberal eras and lose seats in the House during conservative eras. These correlations are weak and not statistically significant. Democratic loss of more seats in Congress in 2004, following their losses in 2002, does provide limited evidence for the idea that the conservative era is still in progress.
Electoral results and party dominance do not correlate well with liberal and conservative eras or with any other cycle. Thus, I have not been able to use the political cycle (or any other cycle) to make valid electoral predictions. As described earlier, a liberal era means that the policies pursued during the post-2001 period will be liberal in comparison to those over 1980-2001. The Bush administration fiscal polices are liberal in that they have reversed the trend to flat or declining domestic spending begun under Reagan. Another liberal trend is the war in Iraq.
President Bush is managing a conflict in Iraq just as his father did. Many of the key players are the same individuals. But the two conflicts have played out very differently. In the first war, the United States was responding to an invasion of Kuwait by Iraq, which cut off Kuwaiti oil from the West. It was clearly an old-fashioned power play by Iraqi leader Saddam Hussein. It was clearly in the interest of the US and the other advanced nations of the world that Iraq should be driven out of Kuwait. A mostly US force was assembled that did just that in a matter of days. President Bush the elder was able to assemble a coalition of rich nations who would largely pay for the cost of the war. The operation was short, relatively inexpensive, and had a tangible payoff--Iraq's attempt to increase its say in the oil market was turned back. It is just the sort of war typically fought during conservative eras.
The second war is entirely different. Beating Iraq in a war is one thing. Occupying the country and installing a new democratic government is a much longer and much more expensive proposition. It was not at all clear how the US, much less the other advanced nations, would benefit from regime change in Iraq. Not surprisingly, President Bush the younger was unable to assemble a similar coalition for the second war. The US proceeded largely on its own, with little financial assistance. Considering that President Bush is a conservative, it is remarkable that his administration has embarked on a program of what conservatives derisively refer to as "nation building".
But nation building is a requirement for installing a democratic government in a country like Iraq. Iraq is rife with ethnic and tribal rivalries and lacks democratic traditions. Much easier would be installation of a brutal dictator like the US did with the Shah of Iran in 1953. Replacement of one murderous tyrant with another is not a politically realistic option today. Today's political zeitgeist reflects the Boomer paradigm of mistrust of government, not the GI paradigm of faith in government that was operative in 1953. Instead, the Bush administration proposes to replace the regime of Saddam Hussein with a democratic government. Before the war, Iraq was said to have large stockpiles of weapons of mass destruction (WMD) which might end up in the hands of terrorists allied with the Iraqi leader. After the invasion, American forces found no WMD. It appears that the potential threat from Iraq was exaggerated to provide a conservative justification for what increasingly looks like an expensive liberal program of nation building.
The obvious question is why would a conservative administration want to engage in such a foreign adventure in the first place? The cycle answer is that we entered a liberal era in 2001 and now large-scale wars for idealistic purposes like "the war to end war" (WW I) become possible. But this tells us nothing about why a conservative like President Bush should start acting liberal simply because a cycle turning point has arrived. Bush has not suddenly become a liberal; he retains his conservative ideology. What has changed is the environment he faces.
The mastermind behind the World Trade Center attack was Osama bin Laden, a Saudi terrorist and Islamic extremist. Bin Laden had issued a call to war (jihad) against the United States largely because he was incensed over the presence of American troops, which he refers to as Crusaders, in the "Land of the Two Holy Shrines" (Saudi Arabia). Other pretexts for war were the US-backed embargo against Iraq, which had caused substantial suffering for the Iraqi people, and the US backing of Israel. The source of most of bin Laden's animus against the U.S. is related to Iraq. Had Iraq not been led by an aggressive leader, there would be no U.S. troops in Saudi Arabia and no Iraqi embargo. There would likely be no jihad against the U.S. and no attack on the World Trade Center.
By removing the troops from Saudi Arabia and ending the Iraq embargo, the Bush administration could reduce a major source of anti-American feeling in the Mideast. Unilateral withdrawal from Saudi Arabia and termination of the Iraq embargo would likely reduce the probability of further terrorist attacks within the US from al Qaeda or like-minded terrorist groups. Just as Soviet withdrawal from Afghanistan ended the jihad against Russia, so would American withdrawal end the jihad against the United States. But this would essentially amount to American surrender to the terrorists. This would be diplomatically unwise as well as a political disaster for the Bush administration. The only acceptable way to accomplish the twin goals of withdrawal from Saudi Arabia and ending the Iraqi embargo would be if the cause of both these policies were to go way. This cause was the intransigent Iraqi regime. The obvious solution was regime change.
Regime change requires that the Hussein government be replaced by something else. The political unacceptability of replacing one dictator with another meant that regime change would have to be accompanied by a program of nation building. And this is what has happened. One can rightly point out that the US presence in Iraq is far more glaring than its presence in Saudi Arabia ever was, and so constitutes a greater provocation for more terrorism. It does. But the US presence in Iraq, unlike that in Saudi Arabia, was not at the request of the indigenous government. The US forces are an army of occupation left by a conqueror. One does not need to be a terrorist criminal to strike at Americans in Iraq, as one had to be to do the same in Saudi Arabia. Hence American forces in Iraq can serve as a "lightening rod", drawing Arab nationalist anger to Iraq (and away from the American homeland). Better the terrorists attempt to engage the heavily-armed US forces in Iraq, who have the authorization and means to fight back, than the US civilian population.
Earlier this year the Iraq war appeared to be shaping into a serious political liability for the Bush administration. The key problem is that Iraqi nation building has no natural support among Bush's conservative core supporters. Conservatives support the war now because they support Bush, not because they like the idea of nation building. The natural supporters for this sort of policy are progressives, who largely are hostile to the Bush administration. A lengthy occupation will create a large population of entitlement-seeking veterans, who will require expensive programs to deal with problems stemming from the war. To deal with them, the Bush administration will likely steer to the left on some issues, while trying to hold a rightward course on others. A good example of this today is the passage of a Medicare prescription drug benefit coupled with right wing judicial appointments. The future could see more welfare spending, especially for veterans, with conservative social policy to try to balance it out.
On the other hand, Bush did win, and a popular vote majority to boot, suggesting that Iraq turned out not to be a political liability after all. With the election now history, there is no reason for the administration to further engage in nation building in Iraq--unless this is truly the objective. Consider how Iraqi disarmament was publicly given as a reason for the invasion, yet when the invasion came, no effort was made to secure Iraqi weapons. Had Iraq actually had WMDs they would have been looted and would now we in terrorist hands along with other conventional arms currently being used by insurgents against coalition forces. Suppose another dictator comes to the fore after the election in January and the US pulls out shortly afterward. This would strongly suggest that nation building was never a goal of the administration, making the Iraqi more consistent with continuation of the Reagan era.
The significance of whether today is in a liberal era comes down to the assignment of the start of a crisis turning. A liberal era beginning in 2001, coupled with the beginning of a secular bear market in 2000, and the dramatic 911 event creates a good case for the start of the crisis turning three years ago. A crisis turning is a time of the unorthodox. Specifically, based on my assessment that a crisis began in 2001, I have been of the opinion that the inflation to come from the twin federal and trade deficits will not result in higher interest rates, which is counterintuitive. The reason is that the last crisis turning (Kondratiev winter) saw tremendous inflationary pressure produced by WW II, and rampant inflation after the war, all associated with very low (and declining) interest rates. This phenomenon represents a change in the way the bond market thinks about interest rates and monetary. This change is called the regime shift, which I have discussed in previous SafeHaven articles [6, 7, 8].
Unfortunately, with the victory of George Bush and the Republicans in the recent election, the case for regime shift in the near future is less certain. This creates an increased likelihood that the crisis start is still in the future, perhaps brought about by a currency collapse? I haven't changed my view to this, but it is still a troubling possibility.
2. Goertzel, Ted, "Generational Cycles in Mass Psychology: Implications for the George W. Bush Administration", International Psychohistorical Association's 24th Annual Convention, June 8, 2001 (www.crab.rutgers.edu/~goertzel/cycles.htm).
6. Alexander Michael A., "The Kondratiev Cycle Revisited: Part One, Current Position in Cycle", SafeHaven, May 5, 2002.
7. Alexander Michael A., "The Kondratiev Cycle Revisited: Part Two, Economic Implications", SafeHaven, May 6, 2002.
8. Alexander Michael A., "The Kondratiev Cycle Revisited: Part Three, Implications for Gold", SafeHaven, May 11, 2002.