I am just a poor boy.
Though my story's seldom told,
I have squandered my resistance
For a pocketful of mumbles,
Such are promises
All lies and jest
Still, a man hears what he wants to hear
And disregards the rest.
– From “The Boxer,” by Paul Simon
In a few hours we will know the outcome of the US elections (hopefully without a repeat of 2000!). So, given that eventuality, why should we bother to explore the rather significant disparity in the models being used to create the polls to predict the outcome of the elections? Because doing so will help us understand why the models we use to predict the effects on our investments of market behavior and macroeconomics so often fail us, and why we should approach the use of such models with a full measure of wariness and skepticism. Yet, at the same time, we should understand when the models may actually be useful, and how to use them.
We all like to think of ourselves as completely rational human beings, ruled by logic and reasonable analysis; but sadly we are not hard-wired for logic. We bring all sorts of biases to our decisions, and even the very way in which we process information in our human brains creates its own set of biases. We are hard-wired to see what we expect to see, and so we not only fail to see the forest for the trees but are equally prone to miss the trees while gazing at the forest. We all too often see what we think we should see (which is one of the reasons it is so hard to edit your own writing!).So, this week, let’s continue our examination of the problems with models, but this time through the lens of political polls.
Speaking of elections, nearly everyone believes that this election will have a major impact on the economy and our investments. And it’s not just about who will be president but also what the US Senate will look like. To address these questions, I am organizing (with a lot of help) a free internet “summit” called “The Post-Election Economy: A Clear-Eyed Analysis of the Risks and Opportunities for Investors.” My friends at Real Clear Politics will be co-hosting with me.
If you're wondering how the election outcome will impact you and your money, we want to help you find someanswers. On November 20th at 2 PM Eastern, we're hosting a free online video event. We'll be bringing in some of the world's leading economics and investing experts, including my friends Mohamed El-Erian, Barry Ritholtz, Richard Yamarone, Gary Schilling, Barry Habib, and James Bianco. Lauren Lyster of RT America will be moderating. And as a very special part of the event, I will be doing a session with the chiefs of staff for both Senator Harry Reid (current Majority Leader) and Senator Rob Portman (one of the true GOP experts on the budget), discussing together the problems we face as a country. These guys get it, and though I will not be asking them to come to definite conclusions or agreement, I think you will come to understand through our discussion that both sides are aware of the seriousness of the situation. (Of course, this session is subject to their last-minute availability.)
This event will be unlike anything you'll get by tuning into the mainstream news. It will be a no-holds-barred discussion on where we can expect the economy and investment markets to go after the election. We'll be discussing the Fiscal Cliff, chronic unemployment and under-employment, our ever-growing deficit and debt issues, over-reaching entitlement programs and their implications, and the impact of QE-Infinity and ongoing easy money policies, as well as risks from Europe, Asia, and beyond.
And most importantly, you'll learn how it all impacts you and your money.
This 90-minute online video event is completely free to attend. You'll be able to watch directly from your home or office computer. All you have to do to register is visit http://www.PostElectionEconomy.com/?ppref=MEC011TF1112A, which I suggest you do now.
I readily admit to being a political junkie, especially at this point in a major election cycle. Although I work hard at not letting my political biases impinge too heavily here, since the driver of the letter is economics and finance, in a former life I was quite involved in political activity, for almost 20 years, getting inside and into the depths of many campaigns over the years. And while I can’t imagine what could get me to once again get that deeply involved, I still maintain an interest. It’s kind of like being a high school athlete – I know what it takes to play the game but have no desire to brutalize myself again.
We have been discussing for the last three weeks how difficult it is to create economic models that accurately describe reality. There are so many assumptions in our models and theories that we have to be very careful in how we use them. Yet politicians and central bankers use these often-too-simplistic models all the time to explain or justify their actual or intended use of various policies in various complex situations. We human beings seem to be attracted to simplistic explanations in the way that moths are attracted to flames, and sometimes with equally disastrous results.
As investors, we want justifications for our decisions. How often do we create models based on past performance and hope – maybe even assume – that the future will turn out that way, too? We look for correlations and trends. We slice and dice the data, hoping to find some direction or insight.
Don’t get me wrong; I use models all the time and will continue to do so. If you understand what a model is actually saying and what its limits are, it can be very useful. Some models can be quite specific and others can give you a sense of direction. Yet, others can be entirely misleading and take you off in the wrong direction or lead you to act too early or too late. Models are like sharp knives – they should be used very carefully and only for the purpose for which they were designed. A paring knife doesn’t work well as a meat cleaver and vice versa. As much attention should be paid to deciding on the appropriate use of a model as to interpreting what it might be saying.
The problems with models apply in the realm of politics, too. We try to discern patterns that can be useful, that tell us how the society around us is thinking.We look for evidence that “our side” is winning. Polls are part of this political modeling process. They are released by campaigns to encourage their supporters and discourage their opponents and to sway those who have not yet made up their minds to join the winning side.
But inside the campaigns, polls are used differently. Not being brutally honest with yourself about what the polls are saying is a sure path to election-night disaster for any candidate or political consultant. You don’t last long in the business of running campaigns if you get a reputation for lying to your clients. You may not want the rest of the world to know the truth, especially if it hurts, but you have to be honest on the deep insides of a campaign.
And in a presidential campaign, too many people eventually know the “inside.” Maybe not the latest polls, but the more recent ones. And given the sheer amount of money being spent on advertising and polls by all sorts of groups, the diversity of opinion about polling results that we’re currently seeing is rather unusual for this late in the election cycle. The following article from Reid Wilson, writing in the National Journal, gives us some insight into the discomfort and confusion inside the campaigns:
“A few days ago, I sat down with Rob Jesmer, the executive director of the National Republican Senatorial Committee. Jesmer is usually tight-fisted about his polling; he doesn't share it with members of the media when the numbers are good for his candidates, and thus avoids the inevitably uncomfortable dilemma when the numbers are bad for his candidates. But he wanted to open his books, if only for a peek, to demonstrate a phenomenon happening across the political spectrum these days: his polls look nothing like polls Democrats are conducting.
“It's a constant refrain from both sides these days. The two parties, the outside groups that are playing such a big role this year, and even some candidates themselves are so dubious about their own numbers that they are employing two pollsters for one race, using one to double-check the other. What flummoxes them even more is that their own party's pollsters are getting similar results, while the other side is offering a completely different take.”
Reid closes with: “Regardless of the cause, strategists on both sides acknowledge the difference in their internal polling. Republicans believe Democrats are counting far too much on low-propensity voters and a booming minority turnout that isn't going to materialize on election day. Democrats believe Republicans are hopelessly reliant on an electorate that looks far more like their party than the nation as a whole. The day after Election Day, somebody's pollsters are going to be proven seriously wrong.
“Deep down, both parties secretly worry it's their side that is missing the boat.”
Last week I mentioned an analysis based on economic data that suggested Romney would win rather handily, along with a few others that drew the opposite conclusion. This week, Nate Silver, writing in the New York Times, offered convincing evidence that Obama will win, and given his generally excellent track record, he is being taken seriously.
Paul Krugman writes that it is not close at all and that Republicans are clearly in panic mode. Yet my experience tells me that you do not decide to spend millions of “last-week dollars” in Pennsylvania if your polls are not telling you the state is in play. Only a few weeks ago, Pennsylvania was a state that was not expected to have any real potential for Romney. Wisconsin, as well. That does not look like panic to me but rather like someone reading a shift in polling data and making a valid effort to push a state into the win column. And yet the Democrats are doing their own polls, too, and those internal polls suggest the Republicans are leaning heavily on misplaced hope and seeing visions in the data that do not jibe with reality.
In a few hours, we will find that the polling professionals on one or both sides were wrong. Contrary to Krugman’s view, it really could be close, and thus the internal polls of both sides may be proven wrong. How can the polls be showing such different outcomes? Reality as we will see it unfold on election night (or the next morning!) will be a very harsh mistress indeed to at least one set of professionals.
It is all about the assumptions. Seemingly minor changes in inputs can have large impacts on polling outcomes. First, remember that polls are conducted by calling an average of 1,000 people and extrapolating to a vastly larger electorate, based on the call results. If 60% of white male voters are saying they will vote for Romney, both campaigns should get roughly that result. But then you have to make an assumption about what percentage of the election-day turnout will be white male voters. If you assume 38%, then Romney gets a certain number of votes, based on your poll. But what if white males are only 34%? Then Romney gets 2.4% fewer votes in your final poll results. And 2.4% can be huge. And what if you assume that the turnout of the black vote, which could easily go 90% for Obama, will be 14% of the total in certain states, rather than 10% – that would be a very large swing. Why would you make either assumption? Maybe because you believe your get-out-the-vote (GOTV) effort will bring in your predicted voters – or you could come up with other justifications for your assumptions.
If you think the electorate will look like it did in 2008, then you’ll get a completely different read from your polling than if the actual voter turnout looks like 2010 or 2004. And of course, as you poll you have to make allowances for people (like me) who no longer have home phones or who work odd hours at part-time jobs.
All those assumptions made inside campaigns are an aside from what you see when you look at the poll numbers on Real Clear Politics. There they have a column right next to the poll estimates for what is called MoE, or margin of error. If the MoE is 4%, then you take the poll number, and anything plus or minus 4% is within a 95% probability of being accurate.
News commentators will from time to time mention the MoE, but seldom do we pay attention to it. We want a simple, unambiguous explanation that we can understand (and I am guilty of that as well!). We want something on which to anchor our emotions and understanding.
It is a very human tendency to see what we expect to see and ignore what we don’t believe to be real. I assume by now you have probably watched the video of a group of people passing a basketball around, even as you failed to notice the person in a gorilla suit walking right across the background! This suggests we might just need to open our minds to new data we were not expecting!
Psychologists have conducted polls in which they presented information on an issue to people who held deeply felt but opposing beliefs about it. Both sides saw the very same data as confirming their own beliefs and biases!
Every high school physics nerd understands the basics of Einstein’s Theory of Relativity. We all did the train-and-platform thought experiment. The answer seems so obvious – that time itself is variable. Albert Michelson and Edward Morley had proved in 1887 that light traveled at a constant speed, but it wasn’t until 18 years later that Einstein puzzled out the “obvious,” because his realization contradicted what everyone knew to be true. No one thought that time could “slow down.”
Let me adapt a few paragraphs relevant to this topic from Bull’s Eye Investing, the book I wrote back in 2003.
People tend to cling tenaciously to their views or forecasts. Once a position has been stated, it is very hard to move away from that view. When movement does occur, it is usually only very grudging. Psychologists call this conservatism bias or confirmation bias (it can lead to anchoring, which I have written about in previous letters).
We see this all the time in earnings forecasts from analysts. They sound nearly religious in their predictions that future earnings will look like past earnings. Analysts are exceptionally good at telling you what has just happened. They have invested too heavily in their views and hence will change them only when presented with indisputable evidence of their falsehood.
Investors should note that these analysts are professionals. We tend to think of them as hard-nosed accountants, sitting around staring fiercely at tables, charts – mind-numbing mounds of data – and arriving at rational conclusions. In reality they are all too human and their humanity shows up all too readily in their forecasts.
This reality leads to a rule:
Don’t get hung up on one technique, tool, approach or view. Flexibility and pragmatism are the order of the day.
We are inclined to look for information we agree with. This thirst for agreement, rather than refutation, is known as confirmation bias. The classic example is the four-card test. Each card carries one alpha-numeric symbol, a set that comprises E, 4, K, and 7. If someone tells you that if a card has a vowel on one side, then it will have an even number on the other, which card(s) do you need to turn over to see whether they are telling the truth?
Most people go for the E and 4 cards. (Confession: I too chose E and 4.) The correct answer is E and 7. Only these two cards are capable of proving the person is lying. If you turn the E over and find an odd number, then the person was a liar; and if you turn the 7 over and find a vowel, then you know they were lying. By turning the 4 over you can prove nothing. If it has a vowel then you have found information that agrees with their statement but doesn’t prove it. If you turn the 4 over and find a consonant, you have proved nothing, since the person stated a vowel card must have an even number on its back, but they didn’t say an even-number card must have a vowel on back!
By picking the 4 card, people are deliberately looking for information that agrees with them. Our natural tendency is to listen to people who agree with us. It feels good to hear our own opinions reflected back. We get those warm, fuzzy feelings of contentment.
These phenomena are all tied up in our human quest for certainty. It is notable that we tend to associate with those who think like we do and confirm the rightness and wisdom of our judgments and views, whether on investments, politics or religion. This behavior only reinforces the tendency to set in concrete wrong views, attitudes … and predictions.
Sadly, this isn’t the best way of making optimal decisions or getting the deepest insights into any topic. Instead of listening to the people who echo our own view, we should make a point to try and understand those who disagree.
Sometimes, if we are not careful, we can create, or buy into, a skewed reality and make bad decisions. How many times have you been told that you just have to buy stocks and hold them for the long run – generally by people who want to sell you their stock funds. And that story has been repeated so often it has become an immutable reality for so many investors, who have now suffered through the last 13 years of bear market.
Lyric Hale brought a Chinese proverb to my attention. It comes from the story of an alleged speech by Pang Cong (龐蔥), an official of the state of Wei in the Warring States Period (475 BC – 221 BC) in Chinese History. According to the Warring States Records, or Zhan Guo Ce, before he left on a trip to the state of Zhao, Pang Cong asked the King of Wei whether he would hypothetically believe in one civilian's report that a tiger was roaming the markets in the capital city, to which the king replied no. Pang Cong asked what the king would think if two people reported the same thing, and the king said he would begin to wonder. Pang Cong then asked, "What if three people all claimed to have seen a tiger there?" The King replied that he would then believe in it. Pang Cong reminded the King that the notion of a live tiger in a crowded market was absurd, yet when the story was repeated by numerous people, it seemed real. Since Pang Cong, as a high-ranking official, had more than three opponents and critics, he was in fact urging the king to pay no attention to those who would spread rumors about him (Pang Cong) while he was away. "I understand," the king replied, and Pang Cong left for Zhao. Yet, slanderous talk took place. When Pang Cong returned to Wei, the King indeed stopped seeing him.
Three men created a tiger in Pang Cong’s life. Be careful of seeing what you want to see in the data you look at. Don’t let your own biases and the problematic use of models create an unwelcome tiger in your life!
I am in Cafayate, Argentina tonight on a beautiful evening in the high plateau country. We came through an amazing canyon to get here, and I am glad I came this way at least once, even if partner and friend Olivier Garret was driving rather aggressively on the winding roads! It is beautiful here, and hopefully time really will slow down over the next few days, as I try to reduce my own relative speed. Thursday afternoon I return to Buenos Aires for a speech the next morning and in time for a historic demonstration that is expected to draw upwards of a million people. That should be fun to watch. I have learned a lot this trip and will write about it next week.
I will go to Chicago, DC, and New York for quick trips prior to coming back home for Thanksgiving. December looks to be a month when I will be home more than usual, and I am happy about that!
Everywhere I go in South America I am asked who I think will win the presidential race. When I say Romney I am met with blank stares and people clearly wondering what planet I am from. When I ask audiences I get a sea of hands, sometimes approaching 90%, that thinkObama will win.
By the way, I was saying Obama would win 30 days ago. And my latest reading of the polls, plus some emails and phone calls, have left me somewhat baffled. While I like to think I can be a cold-blooded observer, I am all too aware that, as Paul Simon says, “A man hears what he wants to hear and disregards the rest.” This is such a different election from anything I have experienced, that it is just too hard to call with a clear head. And there have never been so many polls taken. I talked to one veteran who noted that people in swing states are just not answering the phone. You have to go door to door to do a poll, but then you have to find people at home!
But no matter who wins and who controls the Senate, the focus of the country must turn to how to solve the budget deficit, and soon. That one issue will override all else for a time, and it should.
It is time to hit the send button. It is 10 PM and time for dinner in Argentina. The dinner times are just amazingly late down here! I have to confess that I think I embarrass my hosts when I ask to eat at 8 and then we actually sit down at 9. But friends await! I think Doug Casey and David Galland are around the corner.
Have a great week. I see days at the gym, some massages and great conversations, plus a few books, as well as keeping up with some of the usual sturm und drang.
Your hoping we do not have to be thinking about hanging chads analyst,