Author Archives: Israel Diego

38. When to Rob a Bank

Theis: I argue from an economic perspective that individuals may consider robbing banks because they’re cost-benefit analysis mayadvise them to do so, however it is rarely the case that bank robbery is ever economically feasible.

635612884218706983-1835424490_bank-robbery-facebook-joke

As economists we usually have a different view points about stealing, compared to non-economists. A typical would answer from an economist would be: say we factored in an individual’s level of risk aversion, say 3, in our neat little equation, we may conclude that if an individual’s expected return from stealing exceeds the cost of punishment in fines or even the  income from prison time, then by all means, the consumer should steal because he is welfare maximizing! Ofcourse people would normally site one of the 10 commandments or something as to why stealing is bad, thus justifying that stealing simply morally wrong. However this doesn’t deter everyone from stealing anyways. An old 2012 ‘economist‘ article goes over a nice example using bank robberies.

The story begins with prolific bank robber Willie Sutton, who over a lifetime of robberies was able to earn a hefty sum of $2 million. However bank robbers these days are not nearly as lucky. Using FBI data, they were able to obtain that the average bank haul from a robbery was about $10,025, and the probability of coming out from a successful heist was 90%. This guarantees an expected return of $9,022.5, which seems not too shabby. Although having additional members of your bank heist squad can boost the expected take or increase your probability of a successful heist, having to split the pot makes bank robbery less desirable that it might seem as a one man job. However once factoring the costs for getting caught and charged for bank robbery, the rewards don’t seem appealing anymore.

In the U.S, the punishment for bank robbery varies, but the fines range from $1000 to thousands of dollars, and prison sentences can last from 10 to 25 years. In which case, if we consider the best case scenario of me getting caught and paying a $1000 fine and 10 years in prison, then I might have still gained $9,025 from my heist after subtracting the fee, but 10 years in prison, assuming I could have worked at Starbucks for minimum wage for 10 years, I have forgone about $16,000 a year by working full time, a total of $160,000 over the span of 10 years. The $9000 from my heist simply doesn’t cut it, and I would’ve needed 16 successful heists under my belt to at least break even if I were to get caught and charged with robbery. Hence this demonstrates why economists don’t rob banks, and why U.S law enforcement on bank robberies is pretty successful. Nonetheless bank robbery is only one but many other methods that people cheat/steal, such as corporate embezzlement or medicare fraud. And even still, it may be the case that your potential returns from bank robbery are higher if you actually work for a bank. In the end everyone is mentally calculating expected returns in their head, after factoring in their perceived level of risk aversion, like economists do with cost-benefit analysis, before they make a definitive decision.

37. On the ZLB debate, Why Fiscal Policy doesn’t Cut it

Thesis: On the optimal response for monetary policy when on the ZLB, fiscal policy is constrained in boosting aggregate demand, because of the Fed’s 2% inflation target, and thus in order to for it to work effectively, it would require a ‘regime shift’ to price level targeting.

Summer closing nearer, labor market conditions improving, household purchasing power has increased due to the decline in commodity prices, it won’t be long before interest rates are on the rise and the ZLB becomes a thing of the past, at least for America. David Beckworth seems to agree that as the U.S economy improves, the ZLB debate will become moot. Nonetheless, it is important to keep the ZLB debate in mind, since we still don’t have a consensus amongst economists for the best way to conduct monetary policy at the ZLB. David Beckworth write’s in response to Ben Bernanke on the future of Monetary policy, where Bernanke suggests:

A possible direction of change for the monetary policy framework would be to keep the targets-based approach that I favor, but to change the target. Suggestions that have been made include raising the inflation target, targeting the price level, or targeting some function of nominal GDP… a principal motivation that proponents offer for changing the monetary policy target is to deal more effectively with the zero lower bound on interest rates. But economically, it would be preferable to have more proactive fiscal policies and a more balanced monetary-fiscal mix when interest rates are close to zero. Greater reliance on fiscal policy would probably give better results, and would certainly be easier to explain, than changing the target for monetary policy.

Bernanke takes a stance that a monetary-fiscal mix would’ve been preferable in generating greater aggregate demand over the past six years. However Beckworth argues that fiscal policy is limited by its ability to boost aggregate demand, because of the Fed’s 2% target inflation rate. This is interesting because part of Bernanke’s blog post, suggests that tinkering with the 2% inflation target for future monetary policy would be costly, mainly because monetary policy since the Great Moderation has anchored inflation expectations at 2%, thus changing the level of inflation target  would take time in establishing long-term credibility in order to deviate from the history we have had with 2% inflation targeting. Although Bernanke suggests that the monetary-fiscal mix would be optimal, he acknowledges that  “the probability of getting Congress to accept larger automatic stabilizers.. is low.”

Beckworth demonstrates with an example, that for fiscal policy to have worked in reducing the output gap over the past 6 years, we would’ve needed a monetary policy ‘regime shift’ to price level targeting. In his example, the regime shift would require the price level target to bring back the PCE to its pre-crisis trend path. In order to bring the PCE back to trend, this would require temporary high periods of inflation. This inflation burst “would be the catalyst that spurred robust aggregate demand growth.”
Further in the example we assume the Fed has made QE2 conditional on the PCE returning to its 2002-2008 trend path. In the graph below, we see three different paths for the price level target which are highlighted as three rates for catch-up inflation, 3%, 4%, and a 5%. We see that greater levels of inflation reduce the amount of time it takes to catch up to the previous PCE trend.

catch up scenarios

So in order to get robust aggregate demand growth there needs to be a temporary period of higher inflation. However with the Fed’s 2% inflation target, this would be infeasible, hence rendering fiscal policy ineffective if it is only able to benefit the economy upon reaching the 2% cap on inflation. This topic is good for discussion because it is an example of why fiscal policy may not be very effective during times of economic turmoil. We see the downsides of anchoring inflation at 2% since we limit the alternative methods for a robust recovery, but although probably unlikely, a regime shift could have resulted in robust recovery during the great recession. 

36. (Revised) Inflation Predictions, Elites vs Regular Joes

Thesis: Consumers with higher education and income can predict next year’s inflation rate better than their counterparts, i.e. the average consumer.

Following up from a previous blog post, I made an effort to contrast Michigan’s survey of inflation expectations, the Survey of Professional Forecasters (SPF), and a RW no-change forecast, all forecasting one-year ahead, and I was unable to conclude beyond a reasonable doubt which forecast model would be the best for predicting inflation, however this post revisits this ideology by considering differing levels of income and education of the Michigan survey participants.

As we consider a consumer’s level of income and education, what should we expect to happen to their predictive power of inflation? Some intuition would tell us that as a consumer’s level of income rises, she may have a higher propensity to save relative to consumers with lower income, because she will have more excess funds at her disposal. Naturally, she may invest these funds, and hold a portfolio comprised of stocks, bonds, and other securities, in order to grow her nest egg for retirement, or set money aside to cushion against economic downturns. Thus it would be important for her to frequently keep track of the inflation rate, to make sure that her investment is not eroded by price increases.

Poor-Man-Rich-ManGiven education, it is plausible that higher education on average leads to higher income for the consumer, so inflation expectations would improve based on the theory I motivated above. However we can also infer that more educated consumers are more likely to read the newspaper and be more informed about fluctuations in the inflation rate, a claim made by Christopher Carrol. Lastly it may be the case that less educated consumers fail to muster up an estimate for the inflation rate, relative to higher educated consumers, as uneducated consumers would rather not predict the inflation rate, for fear of embarrassingly overshooting the actual rate. To test this claim, we can sum up the number of survey participants according to their income group, from the Michigan Survey, who did not know what to say when asked about their inflation expectation. We see there is evidence that indeed people with lower education were not able to provide an answer more frequently than individuals with a graduate degree. This is looking at the far ends of the education spectrum, but the relationship holds nonetheless, as we increase education, people less frequently fail to give an answer for their inflation prediction. Interestingly it seems that consumers with less than a high school education know less, on average, about inflation now than they did back in the 80s.

Knowledge of inflation

My claims above would have no ‘oomph’ to them if I didn’t back them up with any evidence. So are we right in assuming that higher education and income lead to better inflation predictions? Intuitively yes, and I provide evidence that this is actually the case.
Let’s take different education and income subgroups, using the Michigan Survey one more time, and lets take the median inflation expectation at each quarter for each subgroup. I choose the median over mean expectations because, this measure is more robust against outliers in the data, and the differences between income and education subgroups are more stark when I use the mean. Finally lets compare each subgroup against one another and see how well they do in comparison at predicting the median-CPI in our experiment. (Reason for this specific inflation measure is explained on last week’s blog). I also use the same Diebold-Mariano test used on my previous post, to test for statistical significance of predictive superiority amongst subgroups. The time period I analyze is 1982-2014, because 1981Q3 is when the SPF began.

Here is a graph comparing graduate level education vs less than high school, and the top 25% income group against the bottom 25%. What we see is that higher education and income tend to move more closely with the Median-CPI. One thing to note is that inflation expectations for low education and income groups are higher  than their counterparts on average for most of the time period analyzed (1982-2014), which provides further intuition that lower educated and poorer consumers not only fail to predict the inflation better than their counter parts, but constantly over predict the rate.

High Inc vs Low Inc

High Educ vs Low Educ

So let’s look at the results! Below I report values for income, as the probability that the highest income group performs better than the alternative income groups. Likewise for education, I display the probability that those with graduate level education do better than their counterparts. Finally we put the highest education and income groups against the SPF, and settle the forecasting battle once and for all. In all of the following results, we should interpret them as probabilities in repeated trials.

Note: These are actually p-values, but I report them this way to make the results more intuitive.

Note: These are actually p-values, but I report them this way to make the results more intuitive.

The above table demonstrates that survey participants grouped in the top 25% bracket, consistently beat their counterparts with about 99% probability. However when the predictions of the top income group were tested against those from the SPF, there was only a 34.31% probability that the high income group would predict better. Similarly, there is large evidence that the group with graduate school education beats every other education group’s predictions with a 99% probability, except for those with a college Bachelor’s degree, but still beating them with a 92.08% probability. The high education group also lost miserably to the SPF, with a probably of fairing better than the SPF of 27.69%. We get very similar, statistically significant results, when we compare the alternative income groups against a lower levels of income and education against the lowest income and education levels. From this we gain two conclusions:

1. As income and education increase, predictive ability for inflation increases unambiguously.
2. There is no subgroup from the Michigan survey that can predict the inflation better than the SPF.

One final note, I find out conclusively that the SPF can do better than all  income groups, although the top 25% group comes closest to the SPF. This result would satisfy the claim in the beginning of the post that those with higher incomes may have a higher predisposition to know the inflation rate because they would be more likely to have an investment portfolio, and it is only those in the top 25% income bracket are able to make predictions somewhat as close to the SPF’s. This claim would be consistent with the fact that about half of Americans hold any assets, hence only individuals in the highest income groups would have any interest in the inflation rate.

36. Google v. Commission

Thesis: Google’s behavior in the European Union, shows that company has been abusing its dominant position, and therefore the probability of a settlement for Google is unfavorable.

googlemonopoly

Back in 1993, Microsoft came under scrutiny for allegedly blocking its competitors out of the market  through antitrust practices. This resulted in the highest-profile antitrust suit between a corporation and the European Commission resulting in Microsoft paying $1.8 billion in fines through 2012. Google is the next tech firm to come under scrutiny for unfairly exercising it’s dominant position (monopoly power in Euro jargon), resulting in other smaller tech firms not being able to compete fairly. Google’s monopoly power is unquestionable, as they handle 90% of web searches in Europe vs 75% in the U.S.

Google is accused of the following anticompetitive tactics:
1. Google favored its own shopping, local, travel, and finance sites over its rivals, resulting in a bias of search results.
2. Copying content from rival sites to improve its own (Scraping).
3. Restricting advertisers’ ability to run campaigns on rival search engines

An interesting note, competition regulators in Europe, more so than in the U.S, are often more concerned with the possibility of mergers distorting competition in the Euro area, otherwise known as a ‘double standard’ in Euro speak. Mr. Döpfner, CEO of Axel Springer SE, was prevented from purchasing ProSiebenSat1 Media AG in 2006 in Germany, but Google on the other hand has acquired over 170 companies during its lifetime, where most of these companies were acquired in the U.S. Thus Google is able to come into the European market with an upper hand. This is a point that Mr. Döpfner makes, that European tech firms are just not able to compete fairly with foreign giants who are allowed to get so big. I consider this as negative publicity for Google, which can only hurt it’s position in court given this stigma from the smaller Euro tech firms.

The EU is in its last stages of investigating Google, and a draft conclusion prepared in March 2013 by the European Commission took the “preliminary view” that Google was abusing its dominant position in all the areas mentioned above. The EU investigation as WSJ article mentions, is actually “part of a broader EU strategy to knit together the bloc’s fragmented online ecosystems into a digital single market. Policy makers hope that will help European Internet firms to build their clout to better compete with U.S. Web giants, like Google and Facebook.” So if the EU plans on building a more united European digital market, then their probing of Google will aim to prove whether Google’s antitrust practices pose as barriers to other tech firms to compete fairly, and thus the ability to unite them.  Finally Margrethe Vestager, the EU’s antitrust chief, will make Google’s case very difficult, because she has a strong preference for legal certainty of “formal charges” in competition cases over negotiated settlements. The fact that she is very adamant about avoiding settlements at any price, means actually reaching a settlement for google highly unlikely.

35. The Banking System as an Ecosystem

Thesis: Thinking of the Banking system as an ecosystem can help in understanding the factors that led to the financial crisis in 2007.

When we think of ecosystems, we imagine a coral reefs, rainforests, or something of the like involving all living organisms being interconnected with each other and the environment. Rarely  do we think of America’s banking system as an ecosystem, but this type of ideology can lead us to an intuitive way of thinking about our banking industry on how to measure systemic risk, like that which we experienced in 2007. An interesting paper, by Halden & May, argue that implementing ideas derived from ecology can lead us to increased stability in the banking system and ultimately minimizing market risk.

Halden & May, in the context of ecology, describe that there is formal proof in model ecosystems, that as the number and strength of interactions among species increased, the more instability there would be in an ecosystem. These models also show that a greater degree of complexity of interaction between species is related to greater instability. This evidence can provide a benchmark model for our banking ecosystem. Since the financial crisis of 2007, there has been increased debate on how to properly assuage systemic/market risk. The analogy then is between multiple banks and how their level of interconnectedness, size, complexity, can affect market stability.

The 2007 crisis was attributed partly from overexuberance within the financial sector itself, this led to growth of derivative markets. This required the need of Arbitrage Pricing Theory (APT), which allowed one to put a price on future risks, and “thus to trade increasingly complex derivative contracts— bundles of assets—with risks apparently decreasing as the bundles grew.” Financial engineers using APT, had to make some underlying assumptions in their pricing methods such as: perfect competition, market liquidity, no-arbitrage and market completeness. These assumptions gave a very misleading account of potential instabilities in markets to financial engineers, because although riskiness seemed to decrease from increasing supply of derivatives, as the number of derivatives grew, so did their complexity, which made the banking system more unstable. This greater instability from complexity is analogous to that in ecosystems, just as it happened in the derivatives market.

Another relation to the ecologist theory, describes that the structure of the banking system can pose a huge problem for systemic risk. For example, given an economic downturn, if large bank is affected largely by a shock to its balance sheet and therefore is deemed insolvent, what is the probability that other banks will be affected by this shock, and hence become insolvent. The structure of the banking system, consists of a few very large “all-purpose banks,” connected to many smaller banks. Thus if a large bank becomes insolvent, this insolvency can spread throughout other smaller banks. This kind of structure though according to the ecological models, is meant to maximize the number of agents affected from an economic downturn. The same intuition applies to the liquidity hoarding that happened during the financial crisis. As large banks hoarded reserves, more banks applied the same strategy, and thus this liquidity hoarding diminished availability of interbank loans.

Lastly, N. Beale shows that excessive homogeneity within a financial systes, i.e. all the banks doing the same thing, “can minimize risk for each individual bank, but maximize the probability of the entire system collapsing.” Diversity across the financial system, became a problem. In the banks in pursuit of diversification, banks’ balance sheets became increasingly homogenous. For example, banks became increasingly reliant on wholesale funding on the liabilities side of the balance sheet;” in structured credit on the assets side of their balance sheet; and managed the resulting risks using the same value-at-risk models.” From a risk minimizing perspective, banks were acting optimally, but since every bank was doing the same thing, the result was an increase systemic risk. Thus we see that interconnectedness in banks, increased complexity, and homogeneity, increased financial instability. Much of the ecological approach for determining systemic risk using these factors have been adopted. The hope is that theory may further be adapted, to ensure another banking crisis doesn’t occur.

34. Inflation Predictions, Elites vs Regular Joes

meThesis: Consumers with higher education and income can predict next year’s inflation rate better than their counterparts, i.e. the average consumer.

On last week’s blog post, in an effort to match up Michigan’s survey of inflation expectations, the Survey of Professional Forecasters (SPF), and a RW no-change forecast, all forecasting one-year ahead, I was unable to conclude beyond a reasonable doubt which forecast model would be the best for predicting inflation, but in this post I hope to redeem myself.

As we consider a consumer’s level of income and education, what should we expect to happen to their predictive power? Some intuition would tell us that as a consumer’s level of income rises, she may have a higher propensity to save, because she will have more excess funds. Naturally, she may invest these funds, and hold a portfolio comprised of stocks, bonds, and other securities, in order to grow her nest egg for retirement, or set money aside to cushion against economic downturns. Thus it would be important for her to frequently keep track of the inflation rate, to make sure that her investment is not eroded by price increases.

Poor-Man-Rich-ManGiven education, it is plausible that higher education on average leads to higher income for the consumer, which goes back to our previous argument on income. However we can also infer that more educated consumers are more likely to read the newspaper and be more informed about fluctuations in the inflation rate, a claim made by Christopher Carrol. Lastly it may be the case that more educated consumers are simply more likely to know what inflation is in the first place, as opposed to their counterparts. To test this claim, we can sum up the number of survey participants according to their income group, from the Michigan Survey, who did not know what to say when asked about their inflation expectation. We see there is evidence that indeed people with lower education were not able to provide an answer more frequently than individuals with a graduate degree. This is looking at the far ends of the education spectrum, but the relationship holds nonetheless, as we increase education, people less frequently fail to give an answer for their inflation prediction. Interestingly it seems that consumers with less than a high school education know less, on average, about inflation now than they did back in the 80s.

Knowledge of inflation

My claims above would have no ‘oomph’ to them if I didn’t back them up with any evidence. So are we right in assuming that higher education and income lead to better inflation predictions? The answer is a resounding yes!
Let’s take different education and income subgroups, using the Michigan Survey one more time, and lets take the median inflation expectation at each quarter for each subgroup. We use Median, because, this measure is more robust against outliers. (Note choosing mean expectation gives weaker results, in that I am able to statistically reject more often). Finally lets compare each subgroup against one another and see how well they do in comparison at predicting the median-CPI in our experiment. (Reason for this specific inflation measure is explained on last week’s blog). I also use the same Diebold-Mariano test used on my previous post, to test for statistical significance of predictive superiority amongst subgroups. The time period I analyze is 1982-2014, because 1981Q3 is when the SPF began.

So let’s look at the results! Below I report values for income, as the probability that the highest income group performs better than the alternative income groups. Likewise for education, I display the probability that those with graduate level education do better than their counterparts. Finally we put the highest education and income groups against the SPF, and settle the forecasting battle once and for all. In all of the following results, we should interpret them as probabilities in repeated trials.

Note: These are actually p-values, but I report them this way to make the results more intuitive.

Note: These are actually p-values, but I report them this way to make the results more intuitive.

The above table demonstrates that survey participants grouped in the top 25% bracket, consistently beat their counterparts with about 99% probability. However when the predictions of the top income group were tested against those from the SPF, there was only a 34.31% probability that the high income group would predict better. Similarly, there is large evidence that the group with graduate school education beats every other education group’s predictions with a 99% probability, except for those with a college Bachelor’s degree, but still beating them with a 92.08% probability. The high education group also lost miserably to the SPF, with a probably of fairing better than the SPF of 27.69%. We get very similar, statistically significant results, when we compare the alternative income groups against a lower levels of income and education against the lowest income and education levels. From this we gain two conclusions:

1. As income and education increase, predictive ability for inflation increases unambiguously.
2. There is no subgroup from the Michigan survey that can predict the inflation better than the SPF, at least to a statistically significant level like.

Here is a graph comparing graduate level education vs less than high school, and the top 25% income group against the bottom 25% (For those more visually inclined 🙂 ) What we see is that the results are consistent with those in the table above. One thing to note is that inflation expectations for low education and income groups are higher  than their counterparts for most of the time period analyzed (1982-2014).
High vs low income High vs low education

One final note, I find out conclusively that the SPF does better than any income group, except the Top 25%, although the SPF probably does better, we cannot tell beyond a reasonable doubt. This result would satisfy the claim in the beginning of the post that those with higher incomes may have a higher predisposition to know the inflation rate,  because it is only those in the top 25% income bracket that are able to make predictions somewhat as close to the SPF’s. This claim would be consistent with the fact that about half of Americans hold any assets, hence only individuals in the highest income groups would have any interest in the inflation rate.

33. The So-Called future of Wall-Street Investing

Thesis: The new style of investment management, using torrents of news and social media data, to make buy or sell decisions on assets, seems like it may fall victim to spurious regression thus providing incorrect signals for investors endorsing this method.

With increased consumer connectivity through social media, it is no surprise to those financially inclined, that anything and everything can affect stock prices, hence why we see such variability in stock returns. The Wall Street Journal last week highlighted a technology company composed of scientists with no finance background who use scientific method-based approaches to investment management.

Their investing process goes like this:

Scientists at Two Sigma program their machines to cull torrents of information from sources like newswires, earnings reports, weather bulletins and Twitter. They write trading algorithms that spot trends in the data, which they may perceive as “buy” or “sell” signals. For example as one of their models automatic sorts through analysts’ research for patterns in a retailer’s performance, simultaneously another would look for clues in Twitter: where it might identify a pattern, such as frequency tweeting about the firm and correlate that with data showing the number of people visiting stores. Then each model produces a trade signal, which is weighed using a risk-management algorithm before finally executing the trade.

Two Sigma co-founder David Siegel, says the firm’s systematic use of artificial intelligence, “represent the future of investment management.”

Is artificial intelligence and use of massive data sets a good thing?

If this is the future of Wall Street trading, then I don’t endorse it. I would agree to a certain extent that analyzing news about an asset, or industry’s expected performance, can alter stock market trading behavior, however when you begin to include data from numerous alternative news, like the ones stated before, you can run into certain problems. Problems such as spurious relationships, that are not new to Economists, but that Scientists may know nothing about. As Ray Dalio, founder of Bridgewater Associates LP, puts it, “such methods risk placing big bets on spurious relationships, a Bridgewater spokesman says.” Although it may not be quite as obvious in this context, spurious relationships tend to give statistically significant coefficients for the regressors in question, although both data sets may have no correlation whatsoever. For example the following are uncorrelated, yet the regressions give statistically significant results: (Source)

1. Regressing US Export Index, on Australian males’ life expectancy
2. US Defense Expenditure, on Population of South African
3. Disney and the U.S. Industrial Sector
4. General Motors and Water

This same process can apply to the use of oceans data, as “professors from University of California, Davis and others, warned on a 2014 research paper of a trend of overfitting in math-based trading by hedge funds and other money managers, in which random correlations are interpreted wrongly as strong relationships.” Their research backs up the claim that the approach of Two Sigma, can often lead to spurious relationships mentioned above. In which case “buy” or “sell” signals would not be reliable at all, and one would be better off using fundamental strategies, such as forecasting dividend growth, and building regressions with macroeconomic predictors that are good determinants of stock returns.

32. The Economics Profession as a market Inefficiency

Thesis: I express my point of view about Economists, despite me being an economist, that Economists are overcompensated, overrepresented outside and inside academia, and contemptuous of other fields of study.

quote1

As economists we are always on the lookout for market inefficiencies, misallocation of resources, inefficient production practices of a firm, inefficiency on the choice of labor-leisure tradeoff of a worker, or an over/underpriced asset. These examples are but a few of the myriad of other market inefficiencies that economists study, and I commend them for that. However all too often I have found myself holding contempt for other fields of study, given my so-called intellectual superiority, or above-average analysis skills with respect to those other fields attempting a similar feat of study. I can generalize this view as a common trait in most economists, although I wouldn’t get much credibility on this statement, given my lack of reputation, unless I were to introduce famous economists who would agree with this statement. Precisely Noah Smith, Paul Krugman, and Crooked Timber acknowledge this sentiment.

Noah Smith, on ‘Why are Economists paid so much,’ points out that economists are often not afraid to express their superiority over sociologists. Economists are also compensated more than other social sciences, probably because economists have superior technical skills. Noah Smith claims this technical skill “is statistics. Economists learn a lot of statistics — much more than anyone else except for applied mathematicians and statisticians.” And his advice to sociologists is to simply “tech-up” and learn statistics, if they want to beat Economists at them their own game. Crooked timber argues, in response to Noah’s blog, that even statisticians get paid less than economists. Statisticians surely could do a similar if not better job at, say consulting a management firm, on the proper course of action over the next year. Thus I have wondered, if we economists are so obsessed with market inefficiencies, why is it that we have not considered whether the overcompensating economists is not an inefficiency in itself? Crooked timber suggests that “much of the assumed authority of economists (just like the authority, in certain policy roles, of international relations scholars like myself), is socially constructed,” and this relates to publicly acknowledged expertise on the topics economists study with respect to rivaling fields. Their claim seems like a plausible explanation for overcompensating economists and evidence of being overrepresented, since Economists hold the authority over our economy’s monetary policy, and to a certain extent, the well-being of our economy.

Lastly Paul Paul Krugman agrees with the claim, that “economists tend to be intellectually arrogant,” because of the highly hierarchical social setup they live in, coupled with steep gradients of prestige, widespread agreement about what constitutes good work and who is doing it. He goes on to argue that an economist builds reputation, not with special connections to others with privilege, or being head of your student body, they gain reputation based on writing clever papers and giving snappy seminar presentations. Indeed economics is built on a hierarchical structure. I would further extend Paul Krugman’s insight, that economists go on judging each other and intellectuals in other fields, based on perceived intellectual competence.

So are we economists full of it? Yeah probably. Are economists overcompensated? I would say so, given the social construct we live in (that economists are superior to other fields of study), thus also why I partially believe Economics could be a market inefficiency. Do I believe that we are actually superior to others? Well no, I believe that economists are good thinkers, and good at thinking outside the box. However we are also somewhat ignorant, in that much of the assumptions we make in our economic models, such as rational expectations, are grounded on consumers being rational, which is something psychology would definitely have a lot of input into whether consumers are indeed rational. But we take this for granted. We also take matters such as racism, and sexism in the economy as granted, yet we fail to understand exactly why, or the reasons why these social constructs exist, yet a sociologist would have a perfect answer to this question. Thus my point of view is that, we must give more credit to those fields that are not so mathematically or statistically ingrained, but can have powerful claims to back up their points of view, that we as economists would not be able to do.

31. Michigan Survey of Inflation Expectations against the SPF

Thesis: Given the Michigan survey of inflation expectations and the Survey of Professional Forecasters, I test the forecast ability of the surveys to predict inflation.

Since the 1970s, macroeconomists had to change the way they do economic analysis. Thanks to Robert Lucas, he showed that one could not predict the effects of a change in macro policy, given only historical data. Economists then understood that expectations, had to be a contributing factor to conducting better monetary policy. Not long after, the Michigan survey of consumer sentiment was born. Here I consider the part of the survey pertaining to inflation expectations, where participants give input on their predictions for the inflation rate for the following year. In essence, I test the forecast ability of the survey to predict “inflation” against other competing models, such as the (SPF) survey of professional forecasters (which began three years after the Michigan survey), and a random-walk no change forecast as a benchmark model.

In a recent blog, I do a similar process in testing the predictive power of futures contracts against a no-change model and a percent spread (future spot price and today’s spot price) model. Although we were able to conclude that futures contracts had the most predictive power, I did not specifically test (nor would I have been able to) if the superiority of the futures relative to the other models was statistically significant. Luckily this is precisely what we can do for testing the forecast ability of the inflation surveys, and the no change model above. (Note that doing this kind of test is very special, given that usually we are only able to do it with survey data).

So we want to test whether the Michigan survey is a good predictor of today’s inflation, and whether or not it is better than the SPF and the no-change. However the measure of inflation is undefined. Survey participants may have an idea of what kind of price changes they are predicting, but we cannot know for sure, because there are so many measures of inflation. Therefore I do a quick analysis of what inflation measure survey participants are predicting. I calculate the PMSE‘s below:
Screen Shot 2015-04-01 at 10.15.49 PMAfter testing for 5 different measures of inflation, choosing the measure that minimizes the PMSE, it seems clear that both the Michigan Survey and the SPF are forecasting the Median CPI. Also plotting a graph of the Michigan Survey and the SPF vs the Median CPI, we see that both surveys do a reasonably good job of predicting inflation. (Note: I have double checked, and made sure that I plotted the inflation expectations with the corresponding CPI, 12 months ahead).

Screen Shot 2015-04-01 at 10.22.19 PM

We can look at this graph and the PMSE from the table above that the SPF may do a slightly better job of predicting the inflation rate, defined by the median CPI. Formally this requires a Diebold-Mariano (DM) test, where I test the null hypothesis that both models have equal predictive power, against the alternative that the SPF is superior. I use the following equation, where the DM statistic follows a standard normal distribution and the values u0t, and u1t, represent

Screen Shot 2015-04-01 at 10.33.22 PMsequences of squared prediction errors, corresponding to the two surveys we are trying to test. Intuitively the DM test, is almost like a t-test, except we are testing whether or not the PMSEs of the survey forecasts are equal.

We arrive at the following conclusion:
Screen Shot 2015-04-01 at 10.43.15 PMThe DM test of SPF vs Michigan Survey, gives a p-value of .3783, and the corresponding p-values for the Michigan Survey and SPF relative to the No-change forecast are .4032 and .4468 respectively. What we conclude for now is that the predictive power of the all the forecasts compared to each other is not statistically significant. What is nice though, is that we were able to conclude with a formal test that we cannot say the SPF is better than the Michigan Survey beyond reasonable doubt, which means there is some merit to the Michigan Survey. This leads me to the next step, that it may be possible to construct a model of survey participants from the Michigan survey that could be superior to the SPF and statistically significant with a DM test. This would lead us to consider demographics of participants such as income, education, and age, and see whether these have substantial effects on people’s ability to predict inflation. I shall consider this question in a future blog post.
(Matlab Code and Data)

30. Futures Contracts as Predictors for Spot Price of Oil

Thesis: Comparing Oil futures contracts against the expected spot price of oil, I find evidence that futures contracts have slightly greater predictive power than a no-change forecast and a popularly used percent spread forecast model, although futures contracts are still not good predictors of the spot price of oil.

Since the creation of oil futures contracts in 1983, oil futures have sometimes been used as an easy benchmark for forecasting future oil spot prices, without the need to do any regression. A futures contract refers to a predetermined price today, at which a transaction for purchasing oil will take place some periods later. In my empirical analysis, I use monthly prices of crude oil futures traded on the NYMEX from http://www.eia.gov/dnav/pet/pet_pri_fut_s1_d.htm. The data begins on March 30th 1983, when crude oil futures were first traded on the NYMEX, and extends through 28 February 2007. In addition I have considered futures contracts with maturity of 1-year ahead (12-monthly periods) and compare their forecasting ability vs two other models: a no-change recursive random walk forecast, and a recursive spread regression model, popularly used by investors.

The futures contract is simply taken to be, the agreed upon crude oil price today, evaluated at the expected spot price, 12 periods from now. This is exhibited by equation 1 below. The no-change forecast on the other hand implies that the expected price of oil in a year, will simply be the same as the spot price today. In other words, to those more economically inclined, this would fit an adaptive expectations framework. The last model is a bit more computational, but not too bad. Equation 2 demonstrates on the left hand side, the log difference of the expected spot price, h periods from today, in our case 12, from today’s spot price. The regressor on the right-hand side demonstrates the log difference between the futures price today, with 12 months maturity, from today’s spot price.

Screen Shot 2015-03-30 at 10.14.11 PM

Equation 1: Futures Contract Forecast                                        Equation 2: Percent spread Forecast

Now for the results, below is a graph I created, demonstrating how the different models compare to the expected spot price. We want to choose the model that comes closest to predicting the spot price of oil, but simply eyeballing the graph below is not sufficient. Luckily, as discussed in class, we can calculate mean squared prediction errors (MSPEs) of all three models by comparing their 12-month ahead forecast, to the spot price that actually occurred, 12 months after the forecast. Then we choose the model that minimizes this value (table below). The intuition behind this is choosing the model that on average predicts the spot price the best.
Screen Shot 2015-03-31 at 1.25.43 AM

From the values above, we conclude that futures contracts seem to perform the best, because they have the lowest MSPE relative to the expected price of oil. This is surprising, because unlike much popular debate, there exists models that usually outperform futures contracts, as demonstrated in (Kilian, Alquist, 2010), yet usually the model is simply the RW no-change forecast. However these results are still consistent, given that the difference between futures contracts and a simple RW no-change forecast, is actually relatively small, yet both MSPEs are considerably high. In other words, neither model has significant predictive ability. This actually teaches us a valuable lesson, that currently there is no model that can actually predict the spot price of oil very well. This is due to the nature of oil prices which are persistent, but also prone to unanticipated shocks in the global supply and demand of crude oil, world economic activity, and political factors, like wars and oil embargos.

Spot price