Canada Backs Out of Kyoto

On Monday, December 12, as an NPR Marketplace piece documents, Canada pulled out of the Kyoto Protocol blaming the lack of participation of the US and China in the accord. Under the Kyoto Protocol, Canada pledged to reduce its CO2 emissions by 6% compared to 1990 levels. In 2011, Canada decided to move the goal post and seek a 17% reduction from 2005 levels by 2020 (read the history here).

Unfortunately, Canada is nowhere close to meeting either of these goals and appears on a future path of increased emissions as long as I would feel comfortable forecasting into the future (attractor forecast above, dashed red line with 98% prediction intervals).

What's interesting is that Canadian CO2 emissions are being driven predominantly by the North American regional system, that is, by the combined productive power of the US and Canada. What is also interesting is that while US emissions have a chance of stabilizing by 2040 (see my forecast here), Canada's emissions do not.

The Canadian government is basically accurate in the reasons given for pulling out of Kyoto. If US energy demand (and thus CO2 emissions) is ever reduced, Canada will just continue exporting whatever hydrocarbons it can produce to China, India or whomever needs oil. The export orientation of Canada's economy is probably the reason that CO2 emissions are unlikely either to be reduced through international protocols or technological change.

US CO2 Emission Reductions Unlikely as a Result of COP17

COP17 (17th Conference of the Parties to the UNFCCC) started today in Durban, South Africa and a salient topic of discussion was what will happen when the Kyoto Protocol 2012 Emission Targets expire next year. Of particular concern is what the US will do. The graphic above is an attractor forecast out to 2050 for US CO2 emissions. It's useful to discuss the forecast in terms of the Kyoto Protocol.

The Kyoto Protocol was initially adopted in December of 1997, went into force in February of 2005 and is scheduled to expire in 2012. The US Executive Branch ratified the protocol but it was never signed by the US Congress. Had the protocol been signed, it would have committed the US to a 7% GHG (Greenhouse Gas) reduction below 1990 emission levels.

In 1990, the US emitted 4992.3 million metric tons of CO2 according to the US EIA (here). In 2006, the US emitted 5981.6 million metric tons of CO2. Current emission levels are lower as a result of the 2007 Financial Crisis. According to the Earth Policy Institute (here):

Between 2007 and 2011, carbon emissions from coal use in the United States dropped 10 percent. During the same period, emissions from oil use dropped 11 percent. In contrast, carbon emissions from natural gas use increased by 6 percent. The net effect of these trends was that U.S. carbon emissions dropped 7 percent in four years. And this is only the beginning.

The initial fall in coal and oil use was triggered by the economic downturn, but now powerful new forces are reducing the use of both. For coal, the dominant force is the Beyond Coal campaign, an impressive national effort coordinated by the Sierra Club involving hundreds of local groups that oppose coal because of its effects on human health.

In other words, it is possible for the US to reduce emission levels by 7%. However, it's important to add that the reduction was the result of the worst financial crash since the Great Depression. My forecast, above, suggests that the probability of getting back to 1990 emission levels is effectively zero for the foreseeable future.

The best short-term forecast for US emissions can be made directly from a simple Impact Model, the type used by the IPCC to create global emission scenarios:

In the Impact model, CO2 emissions are simply a function of production levels, Q -> CO2. Just for round numbers, in 1990 about 1 million metric tons of CO2 were emitted for every 5 trillion dollars of US real GDP. US GDP went from a low of 12.8 trillion US$ in 2009 to about 13.2 trillion US$ in 2011. You can do the math (or look at my GDP forecast here).

The graph above constructs the attractor for US CO2 emissions from the state of the US economy, not just GDP. The results show emissions increasing until at least 2030 before reductions become probable (the dashed lines are the 98% bootstrap confidence intervals). Getting back to 1990 levels are really unlikely for the foreseeable future.

The relationship between CO2 emissions and US GDP should be plain to see from eye balling the historical data. GDP fell during the Financial Crisis and CO2 emissions fell (from the attractor model, you can see that emissions were above their attractor during the bubble). I think US policy makers are keenly aware of this relationship. And, for that reason alone, there is no chance that the US Congress will ever ratify an international protocol limiting GHG emissions. It would mean effectively limiting GDP growth.

In the IPCC scenarios (here) and in the underlying scientific literature, it is typically assumed that reductions will result from technological change (reductions in emission intensity). That issue will have to be dealt with in a later post. The current problems developing a US Solar energy policy (here) suggests to me, at least, that "We Cannot 'Techno-Fix' Our Way to a Sustainable Future."

US Unemployment 2012 and Beyond

Today the US Labor department released the October unemployment rate and it fell from 9.1 to 9 percent (here). President Barack Obama said the improvement was "positive" but that the economy is still growing way too slowly. His hope, of course, is that the unemployment rate will continue dropping right up to the 2012 election.

If President Obama is looking at some readily available forecasts, however, he must be anxious. The Financial Forecast Center (FFC) is forecasting (here) a huge increase in US unemployment for next year (graphic above).

My own forecast (above) shows unemployment stabilizing around 9% for the rest of 2011. Also of interest is that my forecasting model shows that US unemployment is primarily being driven by the world economy and world commodity markets (particularly the price of oil). The FFC forecasts are based on artificial intelligence techniques (here) which do not have an underlying causal model, so it's not entirely clear why they are forecasting such a large increase in unemployment next year. It will be interesting to revisit these forecasts before the 2012 election.

The FFC long-range forecasts (available by subscription) are only made out to 36 months. Commentators are concerned that the US is in for a "New Jobless Era". Jobs have been permanently lost to technology and globalization. The 2000 Financial Crisis swept away all the jobs created by the housing bubble. The jobs created by the bubble may simply never return.

To look at the real long-range, out to 2050, my model predicts an increase in US unemployment that is in line with the FFC forecast, anything from just over 9% to well above 11%. My guess is that the FFC artificial intelligence model is "seeing" the trend happen a little too quickly give the severity of the financial collapse.

We should also keep in mind that the BLS unemployment data is thought to be biased downward (here). Actual unemployment may be above 18% when considering underemployment and discouraged job seekers.

World Impact Forecasts

In the 3F blog, I have been primarily using state space models to make macro forecasts for the world system and countries within that system. The approach is a dynamic realization of ImPACT models developed by the Human Environment Program at Rockefeller University (here) in an article by Waggoner and Ausubel (2002). ImPACT models include the Kaya Identity used by the IPCC and the EIA and the I=PAT identity used for studying population growth impacts (here). Included in the class of ImPACT models is the neoclassical economic growth model (see note below) which has been used by William Nordhaus and Resources for the Future (here) to make climate change forecasts.

Whether or not state space models provide better forecasts than ImPACT models is an open question. The advantage of ImPACT models is that they can be calculated by hand. The disadvantage is that, as explained below, ImPACT models do not include feedback effects.

The directed graph above describes the causality underlying ImPACT models. Under long-run, full employment conditions, population growth (N) leads to greater aggregate production (Q)--more people mean more workers and, as long as the workers are fully employed, more workers mean more output and more demand. Output creates greater energy consumption (E). Energy consumption leads to greater CO2 emissions. And finally, great CO2 emissions leader to increases in global temperature (T).

The extent of these changes depends on the values of the lower case letters, called coefficients or intensive variables (the upper case letters are the extensive variables). In equation form:

T = N*(Q/N)*(E/Q)*(CO2/E)*(T/E) = N*q*e*c*t

where T is global temperature, N is global population, Q is world GDP, E is primary energy consumption, CO2 is global CO2 emissions, q = Q/N per-capita output, e = E/Q energy intensity of production, c = CO2/E carbon intensity of energy and t = T/CO2 is the climate sensitivity to radiative forcing.

The ImPACT formulation is very general. For example, if you think that CO2 emissions have no impact on global temperature, you can set t=0. In other words, if you can provide values for [N,q,e,c,t] then you can make global climate change forecasts using a hand calculator.

There are two problems with such forecasts: (1) you need to come up with reasonable values for population growth and for the other intensive variables (discussed below) and (2) you have to assume that there are no system feedbacks (for example from environmental degradation to population growth or to agricultural production). Said another way, will the intensive variables change over time?

Discussions of energy emissions or global temperature change all seem to rely on the assumption that population growth, energy intensity, and emissions intensity will all decrease over time and decrease enough to make up for increases in per capita GDP (improving standards of living). Generally, the climate sensitivity parameter is assumed to be constant.

Coming up with values for intensive variables in ImPACT models is thus yet another forecasting problem. Below I provide business-as-usual (BAU) forecast for each intensive variable using the WL20 model (data definitions are available in Appendix F). World population forecasts from the United Nations can be found here.

The BAU forecast for real per capita GWP (Gross World Product) is displayed above. Values in 2100 (the usual end-point of long-run forecasts) range anywhere from about 0.02 to 0.09 with a mid-range value of about 0.05 - 0.06.
The BAU forecast for energy intensity in millions of tons of oil equivalent is more difficult to forecast. Energy intensity will not go to zero as predicted by the model so values between 50 and 150 for e = E/Q would seem reasonable.
The BAU forecast for emission intensity (atmospheric CO2 concentrations in ppmv) is displayed above. Values seem to be stabilizing around 0.05.
Finally, BAU forecasts for climate sensitivity (degrees C) provide another surprise. Rather than being a constant as assumed by the IPCC, there would appear to be an observable time trend where climate sensitivity is decreasing. The decrease over time is possibly the result of feedback mechanisms. It is certainly not zero as assumed by global warming skeptics.

The observed trends in energy intensity and climate sensitivity might suggest that using ImPACT models to make long-run projections is an uncertain business. Since imPACT models are identities (true by definition), my suggestion would be to use ImPACT models to make the following types of assertions: (1) if other things remain equal, an increase in population growth would have the following impacts on production, carbon emissions and global temperature or (2) changes in intensive variables necessary to limit global warming to 2 degrees C would involve limiting per capita income growth or reducing energy intensity or decreasing carbon intensity.

NOTE: Neoclassical Economic Growth Models are a form of ImPACT model, which can be demonstrated using directed graphs.

In the standard neoclassical growth model, full employment and growth in autonomous technical change (A) drive growth in output. The Capital Stock, K, is an endogenous variable based on saving from output, K(t) = K(t-w) + ( sQ - d K ) where d is depreciation and s is saving.
Using graph theoretic rules, endogenous variables can be eliminated from the model. Technological change can also be endogenized assuming learning by doing. These two assumptions result in the graph above.

The neoclassical growth model is thus equivalent (nonparametrically) to a dynamic version of the ImPACT model displayed in the graph above.

Iceland's Experiment With Neoliberalism

Paul Krugman recently wrote a NY Times Op Ed piece (here) as a follow up to his earlier "Icelandic Post-crisis Miracle" Op Ed (here). He thinks that Iceland is yet another example of an economy that "produced a decent standard of living for its people" but "was in effect hijacked by a combination of free-market ideology and crony capitalism." Of course, his analysis has been disputed (here and here).

Krugman compared Iceland's percentage changes in GDP since the start of the financial crisis (2007Q4) to Estonia, Ireland and Latvia. He argued that Iceland's approach (let the banks fail and continue deficit spending) allowed it to better weather the crisis than other countries. The critics argue that he was cherry picking starting points for comparisons and if you look back to 2000Q1 or even 2007Q3, Iceland's performance doesn't really look that good.

In some future post I'll look at Estonia, Ireland and Latvia, but for now I would argue that all such comparisons are questionable. Without having an attractor value for GDP growth in each country, it's not really possible to distinguish "bubble" growth from "normal" growth. The graph at the start of this post displays a simple business-as-usual (BAU) attractor path and 98% bootstrap confidence intervals for Iceland's GDP (in real $2000 USD). That graphic tells a better story about Iceland's late 20th century economic history than do the comparisons to other bubble economies provided by the Council on Foreign Relations.

For much of the mid-20th century, the Icelandic economy was underperforming. However, from the late 1980's until the mid-1990's, the economy went through a period of almost stagnant GDP growth. In response, Iceland undertook extensive neoliberal reforms (here). And, in the late 1990's the economy certainly seemed to take off into sustained growth.

Around 2001, however, the economy had reached is BAU attractor and, indeed, there was a period of slow growth. My guess is that growth in the 2% range around the attractor was simply not acceptable and Iceland's answer was financialization which seemed to be producing high growth rates for the foreseeable future (dark red arrow in the graphic) until the Icelandic financial crisis of 2008-2011 brought the economy back to the BAU attractor in 2010.

Compare Iceland to Germany (here). What Iceland has going for itself right now is that (1) it didn't overshoot its BAU attractor by that much (by 2020 the economy will be back to peak bubble GDP values) and (2) it is currently at or near its BAU attractor value (unlike Germany on both counts). The question for Iceland is whether modest growth rates stabilizing around 1% will be acceptable to right-wing, neoliberal policy makers.

Is Another EU Recession Likely?

Concerns are developing (here) that the European sovereign debt crisis could trigger another recession that spreads from the EU area to the US. In an appearance before the US Congress yesterday, US Fed chair Ben Bernanke warned more government action would be needed to prevent a recession in the US. The transmission for this recession would be the banking systems of both the EU and the US which are heavily interconnected.

My business-as-usual (BAU) GDP forecast for the EU is presented above. Actual GDP is displayed as a solid line while the attractor value is the dashed red line with the 98% bootstrap confidence intervals displayed in green and blue dashed lines. After 2003, the EU bubble began developing and peaked in 2007 followed by a crash in 2009 to very low levels in 2010. The model suggests that forces will began pushing the economy back to its attractor value but that does not mean that that future shocks (such as a Greek default) could not push the EU to improbably low GDP levels.

The European Economic Commission's forecast for 2010-2012 (here) suggest that:

The European Commission's autumn forecast foresees a continuation of the economic recovery currently underway in the EU. GDP is projected to grow by around 1.75% in 2010-11 and by around 2% in 2012. A better than expected performance so far this year underpins the significant upward revision to annual growth in 2010 compared to the spring forecast. However, amid a softening global environment and the onset of fiscal consolidation, activity is expected to moderate towards the end of the year and in 2011, but to pick up again in 2012 on the back of strengthening private demand.

The graph above displays the EEC's confidence interval for GDP growth rates. The EEC forecast suggests a small probability of negative growth rates after 2011.
The annualized growth rate of the BAU attractor for GDP in the EU is displayed above. My forecast is for a continually decreasing growth rate approaching zero after 2060.

The definition of economic depression (here) is a little squishy (a drop of more than 10% in GDP lasting for three to four years). In terms of attractor models, the EU economy has been underperforming ever since 2009. The rate of return to the modest growth rates predicted by the BAU GDP attractor will depend on future financial and non-financial shocks to the EU economy.

The German Recession from 2000-2005

The conventional wisdom (here, here, here, and here) is that the German economy was in recession from 2000-2005. Results from the DE20 model, however, show that the economy was well above the attractor (dashed lines above) from 1998-2009 when the recession really hit.

Conventional ideas about dating recessions are essentially based on drawing lines on time plots (the heavy red lines above) rather than modeling an attractor for the economy. The attractor analysis tells an entirely different picture of post-reunification German economic history. The conventional wisdom (here) led to neoliberal reforms in the welfare system and the labor market. The high growth rate after the reforms (2005-2008) was attributed to the success of neoliberalism. The supposed success was short-lived as the economy returned to its attractor. The farther the economy overshoots the attractor value, usually the worse the crash afterwards.

When will there be a Palestinian State and Peace in the Middle East?

The Palestinian authority has approached the United Nations asking for statehood (here). What are the chances the initiative will be successful? What are the chances for peace in the Middle East?

I usually use historical data and statistical models to make predictions. In this case, the cartoon video above does a pretty good job of making the forecast: probably not in my lifetime.

Failed States and The Possible Resurgence of Polio

The Earth Policy Institute has raised the possibility (here) that polio, essentially eradicated in 2000, might spread again from failed states to the rest of the world. From the report:

Once endemic to 125 countries, today polio transmission continues uninterrupted in only 4 countries: Afghanistan, India, Nigeria, and Pakistan; all but India are considered among the world’s top failing states.

Failing states, those that lose control of part or all of their territory and can no longer ensure their people’s security, can pose a threat to international health. They may lack a health care system that is sophisticated enough to participate in the international network that controls the spread of infectious diseases, as illustrated by recent missed opportunities to eradicate polio.

The Earth Policy Institute also contrasted polio to smallpox which the Institute believes has been effectively eradicated. Why smallpox and polio should behave differently in failed states, the article does not say. In a prior post (here), one of my models suggests that smallpox (a disease for which no effective treatment was ever developed) might be linked to trends in commodity markets and might have a resurgence if commodity markets begin malfunctioning.

My models do not show a similar dynamic for polio. The attractor forecast (above, the black line is the actual number of polio cases while the red line is the attractor value and the others are the 98% prediction intervals for the attractor) suggests that, with very high probability, polio was eradicated in 2010.

Has Smallpox Been Eradicated?

The Earth Policy Institute published today (here) an article on smallpox and polio. Smallpox caught my attention after reading the following quote from the article:

Smallpox plagued humanity for thousands of years. In the 18th century, smallpox killed one out of every ten children in France and Sweden. Over the 20th century, the virus caused between 300 and 500 million deaths worldwide. No effective treatment was ever developed.

The eradication of this devastating disease is one of public health’s greatest achievements. It involved mass vaccinations and surveillance to track and contain outbreaks. In 1977, ten years after the World Health Organization (WHO) began an intensive eradication program, the last naturally occurring case of smallpox was identified in Somalia. And on May 8, 1980, the World Health Assembly declared smallpox eradicated.

What also caught my attention was the large spikes in the number of reported cases after World War II, in the late 1950's and in the mid-1970's. My question was what caused the spikes and whether we could be sure that surveillance and vaccination were enough to prevent future outbreaks.

The result of my forecast from the WL2o model is presented in the graphic above. It suggests that future outbreaks are possible. The reason is that the late 20th century spikes are related to spikes in commodity market prices. Whatever the mechanism, future problems in commodity markets (which can be anticipated based on peak oil forecasts) could have unanticipated population health consequences.

Fiction: The Obama GDP Counterfactual

Time magazine recently ran an article titled "The Counterfactual President: Obama Averted Disasters, but Getting Credit Is the Hard Part". The article explains how most of the accomplishments of the Obama administration are based on the counterfactual argument that things would have been worse without the policies enacted by the administration.

This particular quote from the article caught my attention since it implies a testable proposition:

The most extreme example, of course, was the $787 billion stimulus package that Obama signed during his first month in office, when the economy was shedding 700,000 jobs a month. The immediate goal was to avoid a depression, and in that sense it was a tremendous success, stopping the hemorrhaging and stabilizing the scariest economic situation since the Great Depression.

The idea that steps taken by the administration prevented the onset of the second Great Depression implies the following causal model.

The Obama administration, O, after it took office in 2009 not only solved the Subprime Mortgage Crisis, SC (which was negatively affecting the state of the economy, S), but also through the stimulus package improved the state of the economy. Since Gross Domestic product, GDP, is dependent on the state of the economy, improvements in the state of the economy should have improved GDP. To be clear, the model above supposedly explains the actual history we saw after the Obama administration came to office in January of 2009.

The counterfactual argument ("Things would have been worse without us") is described in the model above. The Subprime Mortgage Crisis would have continued to have a negative impact on the state of the economy (and thus GDP) to the present without the policy steps taken by the Obama Administration.

The null hypothesis for the two models above would be that the Subprime Mortgage crisis was over in 2008 and the economy was on the road to recovery in any event. The self-loop in the state of the economy, S, is meant to show that the US economy has its own internal feedback dynamic in response to external shocks such as the Subprime Mortgage Crisis, SC.

A complicated way to test the proposition would be to try to directly estimate the effects of the stimulus on the US economy and demonstrate that the effects were positive. This approach has been tried by the Congressional Budget Office (here), the council of Economic Advisors (here) and economists Alan Blinder and Mark Zandi (here). They have all concluded that the effects were positive.

A simpler way to test the counterfactual would be to estimate a state-space model of the US economy (the USL20 model) and use the state of the US economy to predict real GDP up to the last quarter of 2008, right before the start of the Obama administration. Then simulate the USL2008 model from the first quarter of 2009 until the present. In other words, run the model forward (counterfactually) as if the policies of the Obama administration had never been enacted. If the economy goes off the cliff, than there would be evidence for the counterfactual assertions of the Obama administration ("Things would have been worse without us"). If the economy recovers by itself, then there would be evidence for the null hypothesis ("The subprime Mortgage crisis ended in 2008").

The first question should be how good a job did the USL2008 model do of predicting real GDP (GDP2005 or GDP in 2005 dollars) for the late 20th century. The graphic above shows that the model did a very good job.
The USL2008 model can then be run forward to the present (BEA is currently publishing GDP data into the second quarter of 2011). The model indicates that the economy would not have fallen off a cliff and was in fact underperforming. However, this result cannot be used to argue that the Obama administration policies were inhibiting economic growth (the Republican argument), as the next graphic demonstrates.

There is the matter of probability and the associated prediction intervals for the simulation. As shown above, GDP is still within the 98% prediction interval for this model, but it is getting very close to the lower prediction interval (improbable poor performance) in the second quarter of 2011.

The counterfactual simulation, however, does show that the economy would have hit bottom in mid-2009 regardless of Obama administration policies. The poor performance from mid-2010 to the present could be the result of many other factors (e.g., poor performance in the world economy) other than or in addition to policies of the Obama administration. The simulation, however, does not demonstrate that the "stimulus was a tremendous success" as argued by the Time magazine article.

Forecasting During The Great Recession

Tyler Cowen (here) has argued that since we do not have models that describe the current economic crisis, we really won't be able to forecast when the economy will recover (even though many forecasters are currently trying). Cowen makes a reasonable argument and has certainly thrown down the gauntlet to the forecasting community.

There are a group of forecasters led by the notorious J. Scott Armstrong (Armstrong bet Al Gore $10,000 US that future global temperature cold not be predicted--Gore refused the bet) who think the have the forecasting problem solved. The group argues (here) that there are 139 forecasting principles that are used by successful forecasters. Now would be an interesting time to see whether these principles could improve models of the US economy and lead to better forecasts.

Here are a few random thoughts on current approaches to forecasting:

• The focus should be on our existing models. However, I don't agree with Cowen that a model has to explain everything, for example, business confidence or regulatory uncertainty. A simple canonical forecasting model would be Q(t) = a + b Q(t-1) + e(t), the simple difference equation. A lot of forces such as "confidence" and "uncertainty" must fall into the error term, e(t). It's surprising how many forecasts are presented without prediction intervals based on observed error terms.
• The principles approach to forecast is way too ad hoc. If a forecast doesn't work, there's always a violated principle somewhere to explain the failure. One principle. for example, is that some things (like global warming if you believe Prof. Armstrong) cannot be forecast (the perfect escape clause). However, in that case, the model reduces to Q(t) = Q(t-1) + e(t), the random walk. In another blog (here), I'm searching for random walks in the stock market. Contrary to widely held academic opinion (here), finding stocks that are clearly random walks is not that easy.
• These considerations bring us back to the model again and particularly how the error term, e(t), is treated. One explanation for forecasting failure leading up to the Great Recession is that the Recession itself was a black swan event, unusual and improbable given widely held prejudices about normal distribution theory and error terms.
• There are problems with the way forecasts are typically constructed. If we try to predict tomorrow's Q(t) from yesterday's Q(t-1), we have to remember that Q(t-1) is not itself known with certainty. If errors tend to accumulate over time (as in the random walk model), the system may be very far away from reasonable values (as it was in the Subprime Mortgage crisis). In other words, forecasting during a bubble will work fine until the bubble bursts when the forecasts will be quite wrong.

To summarize, I think Tyler Cowen has it backwards. Forecasting right now should be easy. The bubble has burst. Forecasting when the next bubble will develop (a "return to normalcy") is another problem entirely.

Most of these issues are the subject of this blog and will be the topic of future posts.

About

The title for this blog is taken from an influential book by Nelson Goodman (here) Fact, Fiction and Forecast. In the book, Goodman explores deep issues raised by simple statements: "Global temperature is increasing" (fact, but how do we know this?), "Global temperature would not be increasing without human interventions such as fossil fuel burning" (a fictional counterfactual statement implying a causal link) and "Global temperature will increase above 2 degrees C in 2100" (a forecast, but how can we know the future?)

We don't just have to focus on the Intergovernmental Panel on Climate Change (the IPCC). Consider the current Global Financial Crisis that started in 2007: The financial crisis was caused by a meltdown in the subprime mortgage market (fact?). Had there been stronger financial regulation, the Global Financial Crisis would never have happened (counterfactual?). Had the US government not bailed out the big banks, the financial crisis would have been even deeper (counterfactual). There will continue to be financial crises in the future since financial crises are an inevitable part of the capitalist system (forecast).

Of course, I could go on (and will in this blog)! Nelson Goodman's work has been extended by Judea Pearl (here) in Causality: Models, Reasoning and Inference, by Kevin D. Hoover Causality in Macroeconomics and by others. I hope to discuss their work in future posts.

I also hope to use systems models to establish historical facts, run historical counterfactuals and make forecasts. My results have to be carefully evaluated against the philosophical work on causality: How do I know the historical data I am using is factual (does anyone believe statistics being published by authoritarian regimes)? Are my models good enough to be used for serious historical counterfactuals (just how would that be done)? Finally, how can we really know anything about the future?