What Mr. Trump May Be Thinking About US Trade

It’s often said a few pictures are worth thousands of words. But pictures can be a little misleading, when everything is globally interconnected.

First, look at the standard view of the ever-widening US trade deficit and negative US balance of payments. Then, focusing on Mexico – a Trump whipping boy – let’s pry open the box and look inside to see what is traded back and forth. What the numbers suggest is that the greatest part of the US trade deficit is involved with “trade by related parties”, i.e. multinational companies importing parts and other goods to the US, sometimes for assembly here and export. Many of these are US companies who have used international operations to cut production costs, but then gain access to US customers on favorable terms through their time-honored sales channels.

The Standard Picture

Just looking at the standard US trade statistics, the story is grim. Imports to the US have persistently exceeded US exports since the 1980’s, with the negative balance soaring just before the Great Recession of 2008-2009 to around $200 billion.

Four countries, China, Germany, Mexico, and Japan are the largest contributors to the US trade deficit, as the following chart shows.

Sharp erosion in the US balance of international payments accompanies these import and export curves.

But What Does Mexico Import Into the US?

This is where we have to adopt a new way of looking at these facts.

So, in recent years, US Trade authorities have begun to maintain a new kind of statistical data, relating to trade by related parties, a.k.a. trade between parts of the same (multinational) company.

Mexico, in fact, has a large portion of this trade by related parties, as the following Table indicates.

Readers may also want to consult J.W. Mason’s The Slack Wire What Exactly Does Mexico Export to the US? 

Now there are all sorts of measurement issues involved in measuring trade by related parties, and, of course, the import prices for within-company trade can be somewhat suspect.

But, its interesting some years ago, the Middle Class Political Economist calculated that –

Thus, things are more complicated than suggested by trade in wine from Portugal and textiles from Old Blighty.

In fact, a scholar at Harvard has one of the most compelling pictures highlighting the global supply chain.

Thus, the 787 Development Team encompasses 50 suppliers located in 9 countries (Australia, France, Germany, Italy, Japan, Korea, Sweden, the United Kingdom and the United States). 70 percent of the 787s parts are produced abroad.

 From Pol Antras’ CREI Lectures in Macroeconomics Contracts and the Global Organization of Production, June 2012

So, this is the sort of complexity which enfuriates the “stranded white working class,” left behind when the factories move abroad, but the company marketing organization builds new offices in the nearby metropole.

From which I also deduce that the conflict between Mr. Trump, Mr. Bannon, and powerful interests on the other side is likely to be a serious battle. This is not a win-win, as global trade always was presented (even though it has “distributional impacts”). Border taxes will intervene in these global commodity chains which have been constructed to further the pursuit of profits by multinationals. So border taxes will in fact also have distributional consequences, but these impacts will extend to company profits and involve reorganization of production.

I mean this is like going to be a pitched battle.

What Up? The Trump Years – I

From the standpoint of business forecasting, Donald Trump is important. His challenge to various conventional wisdoms and apparently settled matters raises questions about where things will go in 2017 and beyond. Furthermore, his style of governance is unknown, since as a businessman and minor celebrity, Trump has literally no government experience. He is an Outsider to the political scene, arriving with a portfolio of ideas like mass deportations of illegal immigrants, a massive wall between the United States and its southern neighbor, Mexico, bringing manufacturing jobs back, no further gun controls, and more rigorous screening of immigrants from the Middle East and Muslim countries.

So, with Donald Trump’s Inauguration January 20, a lot seems up in the air. But behind the hoopla, fundamental economic processes and trends are underway. What types of forecasts, therefore, seem reasonable, defensible?

Some Thoughts on Economics

Let’s start with economics.

Donald Trump could be the President who returns inflation and higher interest rates to the equation.

Offshoring and outsourcing have been factors in creating industrial wastelands and hollowed out production in the US – areas where you can drive for miles through abandoned buildings and decaying business centers. At the same time, offshoring and outsourcing bring low-cost electronics and other products to US consumers.

It is a Faustian bargain. If you were a wage-earner with a high school education (or less), supporting a family working in the “fast food sector” or convenience store, maybe holding two jobs to patch together enough income for bills – what you got was a $500 big screen TV and all sorts of gadgets for your kids. You could buy a cheap computer, and cheap clothing, too. Credit cards are available, although less so after 2008.

Oh yeah, another place you could work is in a Big Box store, whose long aisles and vanishing sakes clerks serve as the terminus of global supply chains coursing through ports on the West and East coasts. These are the ports where box-car size containers from China and elsewhere are unloaded, and put on rail cars or moved by truck to stores where consumers can purchase the goods packed in these containers largely on credit.

Is it even possible to slow, stop, or reverse this dynamic?

Let’s see, the plan for bringing manufacturing back to the United States involves deregulation of business, making doing business in the US more profitable. One idea that has been floated is that deregulation would provide incentives for US business to repatriate all that money they are holding overseas back to the US, where it could be invested in America.

Before taking office, President-elect Trump earned points with his supporters by “jawboning” US and foreign companies to keep jobs here, threatening taxes or fees for re-importing stuff to the US from newly relocated operations.

But most of the returning manufacturing would be highly capital intensive (think”robots”) so that only a few more jobs could be garnered from this re-investment in America, right?

Well, before dismissing the idea, note that some of these new jobs would be good-paying, probably requiring higher skills to run more automated production processes.

But this is a different game – producing in the United States, discouraging companies to move operations abroad to lower cost environments, placing taxes, fees, or tariffs on goods manufactured abroad coming into the US. This also involves higher prices.

There is another thread, though, to do with the impact of “deregulation” on the US oil and gas industry, aka “fracking.”

When American ingenuity developed hydraulic fracturing technology (“fracking”) to tap lower yield oil and gas reserves in areas of Texas, South Dakota, and elsewhere, US oil and gas production surged almost to the point of self-sufficiency. But then the Saudi’s lowered the boom, and oil prices dropped, making the higher cost US wells unprofitable, and slowing their expansion.

Before that happened, however, it was apparent fracking in the West and the older oil fields of the Eastern US energized activity up the supply chain, drawing forth significant manufacturing of pipes and equipment. The proverbial boom towns cropped up in the Dakotas and Texas, where hours of work could be long, and pay was good.

Clearly, as oil prices rise again with various global geopolitical instabilities, the US oil and gas industry can rise again, create large numbers of jobs and, also, significant environmental degradation – unless done with high standards for controlling wastes and methane emissions.

But Mr. Trump nominated the former Oklahoma Attorney General to head the US Environmental Protection Agency (EPA) – an agency which Mr. Trump vowed at one point in the campaign to eliminate and which his nominee Scott Pruitt fought tooth and nail in the courts.

So, concern with the environment to the winds, there is a case for a Trump “boom” in 2017 and 2018 – if global oil prices can stay above the breakeven point for US oil and gas production.

Another thread or storyline ties in here  – deportations and stronger controls over illegal immigration.

Again, we have to consider how things are actually made, and we see that, as Anthony Bourdain has noted, many of the restaurant jobs in New York City and other big cities  – in the kitchen especially – are filled by new migrants, many not here legally.

Also, scores of construction jobs in the Rocky Mountain West are filled by Hispanic workers.

Pressure on these working populations to produce their papers can only lead to higher wages and costs, which will be passed along to consumers.

And don’t forget President Trump’s promise to restore US military preparedness. As “cost-plus” contracts, US defense production acts as a conduit for price increases, and may be overpriced (the alternative being to let potential enemies manufacture US weapons).

So what this thought experiment suggests is that, initially, jobs in the Trump era may be boosted by captive or returning manufacturing operations and resumption of the US oil and gas boom – but be accompanied by higher prices. Higher prices also are thematic to limiting the labor pool in US industries, and the cost-overruns are endemic to US defense production.

This is not the end of the economics story, obviously.

The next thing to consider is the US Federal Reserve Bank, which, under Chairman Yellen and other members of the Board of Governors is itching to increase interest rates, as the US economy recovers.

Witnessing a surge of employment from fracking jobs plus a smatter of repatriation of US manufacturing, and the associated higher prices involved with all of this, the Fed should have plenty of excuse to bring interest rates back to historic levels.

Will this truncate the Trump boom?

And what about international response to these developments in the United States?

Daily High and Low Stock Prices – Falling Knives

The mathematics of random walks form the logical underpinning for the dynamics of prices in stock, currency, futures, and commodity markets.

Once you accept this, what is really interesting is to consider departures from random walk movements of prices. Such distortions can signal underlying biases brought to the table by investors and others with influence in these markets.

“Falling knives” may be an example.

A good discussion is presented in Falling Knives: Do Stocks Really Drop 3 Times Faster Than They Rise?

This article in Seeking Alpha is more or less organized around the following chart.


The authors argue this classic chart is really the result of a “Black Swan” event – namely the Great Recession of 2008-2009. Outside of unusual deviations, however, they try to show that “the rate of rallies and pullbacks are approximately equal.”

I’ve been exploring related issues and presently am confident that there are systematic differences in the volatility of high and low prices over a range of time periods.

This seems odd to say, since high and low prices exist within the continuum of prices, their only distinguishing feature being that they are extreme values over the relevant interval – a trading day or collection of trading days.

However, the variance or standard deviation of daily percent changes or rates of change of high and low prices are systematically different for high and low prices in many examples I have seen.

Consider, for example, rates of change of daily high and low prices for the SPY exchange traded fund (ETF) – the security charted in the preceding graph.


This chart shows the standard deviation of daily rates of change of high and low prices for the SPY over rolling annual time windows.

This evidence suggests higher volatility for daily growth or rates of change of low prices is more than something linked just with “Black Swan events.”

Thus, while the largest differences between standard deviations occur in late 2008 through 2009 – precisely the period of the financial crisis – in 2011 and 2012, as well as recently, we see the variance of daily rates of change of low prices significantly higher than those for high prices.

The following chart shows the distribution of these standard deviations of rates of change of daily high and low prices.


You can see the distribution of the daily growth rates for low prices – the blue line – is fatter in a certain sense, with more instances of somewhat greater standard deviations than the daily growth rates for high prices. As a consequence, too, the distribution of the daily growth rates of low prices shows less concentration near the modal value, which is sharply peaked for both curves.

These are not Gaussian or normal distributions, of course. And I find it interesting that the finance literature, despite decades of recognition of these shapes, does not appear to have a consensus on exactly what types of distributions these are. So I am not going to jump in with my two bits worth, although I’ve long thought that these resemble Laplace Distributions.

In any case, what we have here is quite peculiar, and can be replicated for most of the top 100 ETF’s by market capitalization. The standard deviation of rates of change of current low price to previous low prices generally exceeds the standard deviation of rates of change of high prices, similarly computed.

Some of this might be arithmetic, since by definition high prices are greater numerically than low prices, and we are computing rates of change.

However, it’s easy to dispel the idea that this could account for the types of effects seen with SPY and other securities. You can simulate a random walk, for example, and in thousands of replications with positive prices essentially lose any arithmetic effect of this type in noise.

I believe there is more to this, also.

For example, I find evidence that movements of low prices lead movements of high prices over some time frames.

Investor psychology is probably the most likely explanation, although today we have to take into account the “psychology” of robot trading algorithms. Presumeably, these reflect, in some measure, the predispositions of their human creators.

It’s kind of a puzzle.

Top image from SGS Swinger BlogSpot

The Apostle of Negative Interest Rates

Miles Kimball is a Professor at the University of Michigan, and a vocal and prolific proponent of negative interest rates. His Confessions of a Supply-Side Liberal is peppered with posts on the benefits of negative interest rates.

March 2 Even Central Bankers Need Lessons on the Transmission Mechanism for Negative Interest Rates, after words of adoration, takes the Governor of the Bank of England (Mark Carney) to task. Carney’s problem? Carney wrote recently that unless regular households face negative interest rates in their deposit accounts.. negative interest rates only work through the exchange rate channel, which is zero-sum from a global point of view.

Kimball’s argument is a little esoteric, but promotes three ideas.

First, negative interest rates central bank charge member banks on reserves should be passed onto commercial and consumer customers with larger accounts – perhaps with an exemption for smaller checking and savings accounts with, say, less than $1000.

Second, moving toward electronic money in all transactions makes administration of negative interest rates easier and more effective. In that regard, it may be necessary to tax transactions conducted in paper money, if a negative interest rate regime is in force.

Third, impacts on bank profits can be mitigated by providing subsidies to banks in the event the central bank moved into negative interest rate territory.

Fundamentally, Kimball’s view is that.. monetary policy–and full-scale negative interest rate policy in particular–is the primary answer to the problem of insufficient aggregate demand. No need to set inflation targets above zero in order to get the economy moving. Just implement sufficiently negative interest rates and things will rebound quickly.

Kimball’s vulnerability is high mathematical excellence coupled with a casual attitude toward details of actual economic institutions and arrangements.

For example, in his Carney post,  Kimball offers this rather tortured prose under the heading -“Why Wealth Effects Would Be Zero With a Representative Household” –

It is worth clarifying why the wealth effects from interest rate changes would have to be zero if everyone were identical [sic, emphasis mine]. In aggregate, the material balance condition ensures that flow of payments from human and physical capital have not only the same present value but the same time path and stochastic pattern as consumption. Thus–apart from any expansion of the production of the economy as a whole as a result of the change in monetary policy–any effect of interest rate changes on the present value of society’s assets overall is cancelled out by the effect of interest rate changes on the present value of the planned path and pattern of consumption. Of course, what is actually done will be affected by the change in interest rates, but the envelope theorem says that the wealth effects can be calculated based on flow of payments and consumption flows that were planned initially.

That’s in case you worried a regime of -2 percent negative interest rates – which Kimball endorses to bring a speedy end to economic stagnation – could collapse the life insurance industry or wipe out pension funds.

And this paragraph is troubling from another standpoint, since Kimball believes negative interest rates or “monetary policy” can trigger “expansion of the production of the economy as a whole.” So what about those wealth effects?

Indeed, later in the Carney post he writes,

..for any central bank willing to go off the paper standard, there is no limit to how low interest rates can go other than the danger of overheating the economy with too strong an economic recovery. If starting from current conditions, any country can maintain interest rates at -7% or lower for two years without overheating its economy, then I am wrong about the power of negative interest rates. But in fact, I think it will not take that much. -2% would do a great deal of good for the eurozone or Japan, and -4% for a year and a half would probably be enough to do the trick of providing more than enough aggregate demand.

At the end of the Carney post, Kimball links to a list of his and other writings on negative interest rates called How and Why to Eliminate the Zero Lower Bound: A Reader’s Guide. Worth bookmarking.

Here’s a YouTube video.

Although not completely fair, I have to say all this reminds me of a widely-quoted passage from Keynes’ General Theory –

“Practical men who believe themselves to be quite exempt from any intellectual influence, are usually the slaves of some defunct economist. Madmen in authority, who hear voices in the air, are distilling their frenzy from some academic scribbler of a few years back”

Of course, the policy issue behind the spreading adoption of negative interest rates is that the central banks of the world are, in many countries, at the zero bound already. Thus, unless central banks can move into negative interest rate territory, governments are more or less “out of ammunition” when it comes to combatting the next recession – assuming, of course, that political alignments currently favoring austerity over infrastructure investment and so forth, are still in control.

The problem I have might be posed as one of “complexity theory.”

I myself have spent hours pouring over optimal control models of consumption  and dynamic general equilibrium. This stuff is so rarified and intellectually challenging, really, that it produces a mindset that suggests mastery of Portryagin’s maximum principle in a multi-equation setup means you have something relevant to say about real economic affairs. In fact, this may be doubtful, especially when the linkages between organizations are so complex, especially dynamically.

The problem, indeed, may be institutional but from a different angle. Economics departments in universities have, as their main consumer, business school students. So economists have to offer something different.

One would hope machine learning, Big Data, and the new predictive analytics, framed along the lines outlined by Hal Varian and others, could provide an alternative paradigm for economists – possibly rescuing them from reliance on adjusting one number in equations that are stripped of the real, concrete details of economic linkages.

Negative Interest Rates

What are we to make of negative interest rates?

Burton Malkiel (Princeton) writes in the Library of Economics and Liberty that The rate of interest measures the percentage reward a lender receives for deferring the consumption of resources until a future date. Correspondingly, it measures the price a borrower pays to have resources now.

So, in a topsy-turvy world, negative interest rates might measure the penalty a lender receives for delaying consumption of resources to some future date from a more near-term date, or from now.

This is more or less the idea of this unconventional monetary policy, now taking hold in the environs of the European and Japanese Central Banks, and possibly spreading sometime soon to your local financial institution. Thus, one of the strange features of business behavior since the Great Recession of 2008-2009 has been the hoarding of cash either in the form of retained corporate earnings or excess bank reserves.

So, in practical terms, a negative interest rate flips the relation between depositors and banks.

With negative interest rates, instead of receiving money on deposits, depositors must pay regular sums, based on the size of their deposits, to keep their money with the bank.

“If rates go negative, consumer deposit rates go to zero and PNC would charge fees on accounts.”

The Bank of Japan, the European Central Bank and several smaller European authorities have ventured into this once-uncharted territory recently.

Bloomberg QuickTake on negative interest rates

The Bank of Japan surprised markets Jan. 29 by adopting a negative interest-rate strategy. The move came 1 1/2 years after the European Central Bank became the first major central bank to venture below zero. With the fallout limited so far, policy makers are more willing to accept sub-zero rates. The ECB cut a key rate further into negative territory Dec. 3, even though President Mario Draghi earlier said it had hit the “lower bound.” It now charges banks 0.3 percent to hold their cash overnight. Sweden also has negative rates, Denmark used them to protect its currency’s peg to the euro and Switzerland moved its deposit rate below zero for the first time since the 1970s. Since central banks provide a benchmark for all borrowing costs, negative rates spread to a range of fixed-income securities. By the end of 2015, about a third of the debt issued by euro zone governments had negative yields. That means investors holding to maturity won’t get all their money back. Banks have been reluctant to pass on negative rates for fear of losing customers, though Julius Baer began to charge large depositors.

These developments have triggered significant criticism and concern in the financial community.

Japan’s Negative Interest Rates Are Even Crazier Than They Sound

The Japanese government got paid to borrow money for a decade for the first time, selling 2.2 trillion yen ($19.5 billion) of the debt at an average yield of minus 0.024 percent on Tuesday…

The central bank buys as much as 12 trillion yen of the nation’s government debt a month…

Life insurance companies, for instance, take in premiums today and invest them to be able to cover their obligations when policyholders eventually die. They price their policies on the assumption of a mid-single-digit positive return on their bond portfolios. Turn that return negative and all of a sudden the world’s life insurers are either unprofitable or insolvent. And that’s a big industry.

Pension funds, meanwhile, operate the same way, taking in and investing contributions against future obligations. Many US pension plans are already borderline broke, and in a NIRP environment they’ll suffer a mass extinction. Again, big industry, many employees, huge potential impact on both Wall Street and Main Street.

It has to be noted, however, that real (or inflation-adjusted) interest rates have gone below zero already for certain asset classes. Thus, a highlight of the Bank of England Study on negative interest rates circa 2013 is this chart, showing the emergence of negative real interest rates.


Are these developments the canary in the mine?

We really need some theoretical analysis from the economics community – perspectives that encompass developments like the advent of China as a major player in world markets and patterns of debt expansion and servicing in the older industrial nations.

The Arc Sine Law and Competitions

There is a topic I think you can call the “structure of randomness.” Power laws are included, as are various “arcsine laws” governing the probability of leads and changes in scores in competitive games and, of course, in winnings from gambling.

I ran onto a recent article showing how basketball scores follow arcsine laws.

Safe Leads and Lead Changes in Competitive Team Sports is based on comprehensive data from league games over several seasons in the National Basketball Association (NBA).

“..we find that many …statistical properties are explained by modeling the evolution of the lead time X as a simple random walk. More strikingly, seemingly unrelated properties of lead statistics, specifically, the distribution of the times t: (i) for which one team is leading..(ii) for the last lead change..(and (iii) when the maximal lead occurs, are all described by the ..celebrated arcsine law..”

The chart below shows the arcsine probability distribution function (PDF). This probability curve is almost the opposite or reverse of the widely known normal probability distribution. Instead of a bell-shape with a maximum probability in the middle, the arcsine distribution has the unusual property that probabilities are greatest at the lower and upper bounds of the range. Of course, what makes both curves probability distributions is that the area they span adds up to 1.


So, apparently, the distribution of time that a basketball team holds a lead in a basketball game is well-described by the arcsine distribution. This means lead changes are most likely at the beginning and end of the game, and least likely in the middle.

An earlier piece in the Financial Analysts Journal (The Arc Sine Law and the Treasure Bill Futures Market) notes,

..when two sports teams play, even though they have equal ability, the arc sine law dictates that one team will probably be in the lead most of the game. But the law also says that games with a close final score are surprisingly likely to be “last minute, come from behind” affairs, in which the ultimate winner trailed for most of the game..[Thus] over a series of games in which close final scores are common, one team could easily achieve a string of several last minute victories. The coach of such a team might be credited with being brilliantly talented, for having created a “second half” team..[although] there is a good possibility that he owes his success to chance.

There is nice mathematics underlying all this.

The name “arc sine distribution” derives from the integration of the PDF in the chart – a PDF which has the formula –

f(x) = 1/(π (x(1-x).5)

Here, the integral of f(x) yields the cumulative distribution function F(x) and involves an arcsine function,

F(x) = 2/(π arcsin(x.5))

Fundamentally, the arcsine law relates to processes where there are probabilities of winning and losing in sequential trials. The PDF follows from the application of Stirling’s formula to estimate expressions with factorials, such as the combination of p+q things taken p at a time, which quickly becomes computationally cumbersome as p+q increases in size.

There is probably no better introduction to the relevant mathematics than Feller’s exposition in his classic An Introduction to Probability Theory and Its Applications, Volume I.

Feller had an unusual ability to write lucidly about mathematics. His Chapter III “Fluctuations in Coin Tossing and Random Walks” in IPTAIA is remarkable, as I have again convinced myself by returning to study it again.


He starts out this Chapter III with comments:

We shall encounter theoretical conclusions which not only are unexpected but actually come as a shock to intuition and common sense. They will reveal that commonly accepted motions concerning chance fluctuations are without foundation and that the implications of the law of large numbers are widely misconstrued. For example, in various applications it is assumed that observations on an individual coin-tossing game during a long time interval will yield the same statistical characteristics as the observation of the results of a huge number of independent games at one given instant. This is not so..

Most pointedly, for example, “contrary to popular opinion, it is quite likely that in a long coin-tossing game one of the players remains practically the whole time on the winning side, the other on the losing side.”

The same underlying mathematics produces the Ballot Theorem, which states the chances a candidate will be ahead from an early point in vote counting, based on the final number of votes for that candidate.

This application, of course, comes very much to the fore in TV coverage of the results of on-going primaries at the present time. CNN’s initial announcement, for example, that Bernie Sanders beat Hillary Clinton in the New Hampshire primary came when less than half the precincts had reported in their vote totals.

In returning to Feller’s Volume 1, I recommend something like Sholmo Sternberg’s Lecture 8. If you read Feller, you have to be prepared to make little derivations to see the links between formulas. Sternberg cleared up some puzzles for me, which, alas, otherwise might have absorbed hours of my time.

The arc sine law may be significant for social and economic inequality, which perhaps can be considered in another post.

Business Forecasting – Practical Problems and Solutions

Forecasts in business are unavoidable, since decisions must be made for annual budgets and shorter term operational plans, and investments must be made.

And regardless of approach, practical problems arise.

For example, should output from formal algorithms be massaged, so final numbers include judgmental revisions? What about error metrics? Is the mean absolute percent error (MAPE) best, because everybody is familiar with percents? What are plus’es and minus’es of various forecast error metrics? And, organizationally, where should forecasting teams sit – marketing, production, finance, or maybe in a free-standing unit?

The editors of Business Forecasting – Practical Problems and Solutions integrate dozens of selections to focus on these and other practical forecasting questions.

Here are some highlights.

In my experience, many corporate managers, even VP’s and executives, understand surprisingly little about fitting models to data.

So guidelines for reporting results are important.

In “Dos and Don’ts of Forecast Accuracy Measurement: A Tutorial,” Len Tashman advises “distinguish in-sample from out-of-sample accuracy,” calling it “the most basic issue.”

The acid test is how well the forecast model does “out-of-sample.” Holdout samples and cross-validation simulate how the forecast model will perform going forward. “If your average error in-sample is found to be 10%, it is very probable that forecast errors will average substantially more than 10%.” That’s because model parameters are calibrated to the sample over which they are estimated. There is a whole discussion of “over-fitting,” R2, and model complexity hinging on similar issues. Don’t fool yourself. Try to find ways to test your forecast model on out-of-sample data.

The discussion of fitting models when there is “extreme seasonality” broke new ground for me. In retail forecasting, there might be a toy or product that sells only at Christmastime. Demand is highly intermittent. As Udo Sglavo reveals, one solution is “time compression.” Collapse the time series data into two periods – the holiday season and the rest of the year. Then, the on-off characteristics of sales can be more adequately modeled. Clever.

John Mello’s “The Impact of Sales Forecast Game Playing on Supply Chains,” is probably destined to be a kind of classic, since it rolls up a lot of what we have all heard and observed about strategic behavior vis a vis forecasts.

Mello describes stratagems including

  • Enforcing – maintaining a higher forecast than actually anticipated, to keep forecasts in line with goals
  • Filtering – changing forecasts to reflect product on hand for sale
  • Hedging – overestimating sales to garner more product or production capability
  • Sandbagging – underestimating sales to set expectations lower than actually anticipated demand
  • Second-guessing – changing forecasts to reflect instinct or intuition
  • Spinning – manipulating forecasts to get favorable reactions from individuals or departments in the organization
  • Withholding – refusing to share current sales information

I’ve seen “sand-bagging” at work, when the salesforce is allowed to generate the forecasts, setting expectations for future sales lower than should, objectively, be the case. Purely by coincidence, of course, sales quotas are then easier to meet and bonuses easier to achieve.

I’ve always wondered why Gonik’s system, mentioned in an accompanying article by Michael Gilliland on the “Role of the Sales Force in Forecasting,” is not deployed more often. Gonik, in a classic article in the Harvard Business Review, ties sales bonuses jointly to the level of sales that are forecast by the field, and also to how well actual sales match the forecasts that were made. It literally provides incentives for field sales staff to come up with their best, objective estimate of sales in the coming period. (See Sales Forecasts and Incentives)

Finally, Larry Lapide’s “Where Should the Forecasting Function Reside?” asks a really good question.

The following graphic (apologies for the scan reproduction) summarizes some of his key points.


There is no fixed answer, Lapide provides a list of things to consider for each organization.

This book is a good accompaniment for Rob Hyndman and George Athanasopoulos’s online Forecasting: Principles and Practice.

Texas Manufacturing Shows Steep Declines

The Dallas Federal Reserve Bank highlights the impact of continuing declines in oil prices in their latest monthly Texas Manufacturing Outlook Survey:

Texas factory activity fell sharply in January, according to business executives responding to the Texas Manufacturing Outlook Survey. The production index—a key measure of state manufacturing conditions—dropped 23 points, from 12.7 to -10.2, suggesting output declined this month after growing throughout fourth quarter 2015.

Other indexes of current manufacturing activity also indicated contraction in January. The survey’s demand measures—the new orders index and the growth rate of orders index—led the falloff in production with negative readings last month, and these indexes pushed further negative in January. The new orders index edged down to -9.2, and the growth rate of orders index fell to -17.5, its lowest level in a year. The capacity utilization index fell 15 points from 8.1 to -7, and the shipments index also posted a double-digit decline into negative territory, coming in at -11.

Perceptions of broader business conditions weakened markedly in January. The general business activity and company outlook indexes fell to their lowest readings since April 2009, when Texas was in recession. The general business activity index fell 13 points to -34.6, and the company outlook index slipped to -19.5.

Here is a chart showing the Texas monthly manufacturing index.


The logical follow-on question is raised by James Hamilton – Can lower oil prices cause a recession?

Hamilton cites an NBER (National Bureau of Economic Research) paper – Geographic Dispersion of Economic Shocks: Evidence from the Fracking Revolution – which estimates jobs from fracking (hydraulic fracturing of oil deposits) resulted in more than 700,000 US jobs 2008-2009, resulting in an 0.5 percent decrease in the unemployment rate during that dire time.

Obviously, the whole thing works in reverse, too.

Eight states with a high concentration of energy-related jobs – including Texas and North Dakota – have experienced major impacts in terms of employment and tax revenues. See “Plunging oil prices: a boost for the U.S. economy, a jolt for Texas”.

Another question is how long can US-based producers hold out financially, as the price of crude continues to spiral down? See Half of U.S. Fracking Industry Could Go Bankrupt as Oil Prices Continue to Fall.

I’ve seen some talk that problems in the oil patch may play a role analogous to sub-prime mortgages during the last economic contraction.

In terms of geopolitics, there is evidence the Saudi’s, who dominate OPEC, triggered the price decline by refusing to limit production from their fields.

Is the Economy Moving Toward Recession?

Generally, a recession occurs when real, or inflation-adjusted Gross Domestic Product (GDP) shows negative growth for at least two consecutive quarters. But GDP estimates are available only at a lag, so it’s possible for a recession to be underway without confirmation from the national statistics.

Bottom line – go to the US Bureau of Economics Analysis website, click on the “National” tab, and you can get the latest official GDP estimates. Today, (January 25, 2016) this box announces “3rd Quarter 2015 GDP,” and we must wait until January 29th for “advance numbers” on the fourth quarter 2015 – numbers to be revised perhaps twice in two later monthly releases.

This means higher frequency data must be deployed for real-time information about GDP growth. And while there are many places with whole bunches of charts, what we really want is systematic analysis, or nowcasting.

A couple of initiatives at nowcasting US real GDP show that, as of December 2015, a recession is not underway, although the indications are growth is below trend and may be slowing.

This information comes from research departments of the US Federal Reserve Bank – the Chicago Fed National Activity Index (CFNAI) and the Federal Reserve Bank of Atlanta GDPNow model.


The Chicago Fed National Activity Index (CFNAI) for December 2015, released January 22nd, shows an improvement over November. The CFNAI moved –0.22 in December, up from –0.36 in November, and, in the big picture (see below) this number does not signal recession.


The index is a weighted average of 85 existing monthly indicators of national economic activity from four general categories – production and income; employment, unemployment, and hours; personal consumption and housing; and sales, orders, and inventories.

It’s built – with Big Data techniques, incidentally- to have an average value of zero and a standard deviation of one.

Since economic activity trends up over time, generally, the zero for the CFNAI actually indicates growth above trend, while a negative index indicates growth below trend.

Recession levels are lower than the December 2015 number – probably starting around -0.7.

GDPNow Model

The GDPNow Model is developed at the Federal Reserve bank of Atlanta.

On January 20, the GDPNow site announced,

The GDPNow model forecast for real GDP growth (seasonally adjusted annual rate) in the fourth quarter of 2015 is 0.7 percent on January 20, up from 0.6 percent on January 15. The forecasts for fourth quarter real consumer spending growth and real residential investment growth each increased slightly after this morning’s Consumer Price Index release from the U.S. Bureau of Labor Statistics and the report on new residential construction from the U.S. Census Bureau.

The chart accompanying this accouncement shows a somewhat less sanguine possibility – namely that consensus estimates and the output of the GDPNow model have been on a downward trend if you look at things back to September 2015.


Sales and new product forecasting in data-limited (real world) contexts