Critical Analytics: It’s Stories All the Way Down

I’ve been hearing much lately about stories, narratives, analytics, data, and “big data”.  I have no need to call out exactly who or which pieces of writing. You know who you are. My aim here is not to criticize, oppose, or take sides. It’s to take a brief critical look at what’s being discussed.

Much of the discussion strikes me as one tribe (I’ll call them non-quants) pleading that stories and narratives are important too!  All of which is an understandable reaction to how the other tribe (I’ll call them quants) have seemingly gained a favored position and perceived superiority at divining the “truth” because they are evidence based!  Because data! I’m actually a member of both tribes and find the posturing of stories and narratives as alternative to quantitative analysis disheartening.

The most encouraging blog piece I’ve read recently comes from Michael Feldstein.  In his lengthy (and excellent) post called Analytics Literacy is a Major Limiter of Edtech Growth.  Please do read it.   He argues for the dissolving this false juxtaposition between “stories” and “data”.

…some of these arguments position analytics in opposition to narratives. That part is not right. Analytics are narratives. They are stories that we tell, or that machines tell, in order to make meaning out of data points. The problem is that most of us aren’t especially literate in this kind of narrative and don’t know how to critique it well.

I wholeheartedly agree.  Feldstein is (correctly) arguing that data points are nothing without stories.  The meaning we take from the data is itself nothing but a story we weave using the data points as we might use punctuation or particular words.  In essence, quantitative analysis is itself a story.

This really isn’t news or at least it shouldn’t be.  I remember how powerful McCloskey’s Rhetoric of Economics was for me when I read it decades ago.  McCloskey powerfully made the point that no matter how much we wrapped an idea in data, mathematical formalism, or econometric analysis, everything we said in economics was just a metaphor or a story we imposed on the data. Alan Grossman long ago pointed out that even that high temple of data-driven evidence, Science(tm), it’s still just rhetoric and it’s still just stories.

Yes, the meaning we attach to a set of data is itself a story.  So stories are not alternatives to data. Data is a story.  But it’s not just the obvious story we tell with the data. There’s a story unstated underneath the data the we use. Our choice of particular data variables constitutes a story itself. We (or at least the data collector) have in mind a story and narrative of what’s important before they collect the data.  They don’t collect data about the context that they don’t see as important or relevant (or easy enough to collect), so they assume a story about that uncollected contextual data holds no meaning.  There’s a story underneath the story we told with the data.

But it keeps getting deeper. Much like the philosophical turtles, it’s stories all the way down. That measure of the data you’re using. The one you think is just basic stats or math, something like the average (properly called arithmetic mean), or the variance, or correlation, or whatever.  It has a story too.  Let’s take that arithmetic mean (average) and each observation’s difference from the average. We think of that average as “the norm” – but that’s just a story invented by a couple of different statisticians in the 19th century.

I can’t really do justice here to the story of how that story of what the average or norm is.  I strongly urge you to read The End of Average by Todd Rose.  It’s fully accessible to members of both tribes, quants and non-quants.  You’ll never use your quantitative data the same way again. Todd Quinn writing in the Elearning magazine of the ACM had the same kind of dramatic reaction as I had.

I’ve finished reading Todd Rose’s The End of Average, and I have to say it was transformative in ways that few books are. I read a fair bit, and sometimes what I read adds some nuance to my thinking, and other times I think the books could stand to extend their own nuances. Few books fundamentally make me “think different,” but The End of Average was one that did, and I believe it has important implications for learning and business.

Rose’s point is pretty simple: All our efforts to try to categorize people on a dimension like GPA or SAT or IQ are, essentially, nonsensical.

But going another level down, as Rose explains in End of Average, there are assumptions beneath the calculation and use of ordinary stats like the average or the variance.  Let’s face it, “assumptions” is another way of staying “believed a story to be so true that it didn’t need to be stated”.  In the case of the average and the calculation of differences from “the norm”, that assumed story has to do with the ergodic properties of what’s being examined.  So what’s “ergodic  properties”? Well here’s Wikipedia’s attempt to explain ergodicity. It’s not very accessible to non-quants (or even most quants!).  Again, I would refer you  to Rose’s book for a beginning glimpse of what ergodicity means. I can’t explain it here, but the essence is that mathematically, statistically the vast majority of the stories being told with quantitative analytics are complete nonsense. Garbage. Invalid. Wishful alchemy.

It’s stories all the way down.  At first this might seem discouraging. But it’s not. I’m calling for not just analytics literacy but a critical analytics.  We need to investigate and become aware of not only the stories we tell using data, but also the assumed stories we slide under the table by choosing particular measures and statistical techniques without thinking about them. We wouldn’t let the semantics of narratives escape critical examination. Why should we let analytics?

 

Data and Visualization Resources for Incomes and Inequality

Posting links to two incredibly useful resources for students and people doing research on incomes, income distribution, and income inequality. These resources are useful for both historical data and visualizations as well as cross-country comparisons.

The first is the World Top Incomes Database from the Paris School of Economics. Many thanks to the Paris School and researchers Facundo Alvaredo, Tony Atkinson, Thomas Piketty and Emmanuel Saez. It’s a a tremendous resource.

The second is a tremendous resource also. It’s Our World In Data. It’s a work in progress project by Max Roser,   but it’s already jam packed with great data and visualizations on incomes, health, war and violence, poverty, and food and hunger. And best of all, it’s all CC-BY-SA licensed.  I love it when collaboration and the commons come together to support learning.

China, Growth, and the Weakness of Real GDP

Sara Hsu asks if All Growth is Good? The Case of China Of course, not all growth is good. It makes little difference, whether it’s economic or human tissue growth. Edward Abbey famously wrote that “growth for the sake of growth is the ideology of the cancer cell”. Obesity is another form of high-growth, yet it hardly improves well-being or health.

Unfortunately, we economists have not (yet?) developed measures that help us or policy-makers distinguish between healthy growth and malignant growth.  The only real comprehensive measure of growth we have is growth of real GDP. We do know better, as Sara notes:

Since the seventies, with the assertion by Gunnar Myrdal that economic development should prioritize equality, economists have increasingly come to believe that not all types of growth are wholly “good.” Growth that ignores human well-being and equality are viewed as problematic.  Certainly growth that results in severe environmental destruction, as in the case of China over the past twenty years, cannot be classified as good, either, despite the country’s much-lauded successes during this period. Real-world views of growth depicted in the mainstream media do not fall in line, however, with the economic development literature. The focus on China’s growth in the news has distracted from a more balanced view of the looming inequality problems or polluting production methods in the world’s most populous nation.  As China’s growth has slowed, headlines have read, “China’s Economic Growth at Stake,” “China’s Economic Growth Slows,” and “China’s Second Quarter Growth Slows.” –

Yes, China’s real GDP growth rate has been spectacular for several decades now. That growth has lifted literally hundreds of millions of people into better lives. Yet, in strange case of the metaphor becoming real, that economic growth has literally brought cancer with it. Specifically, many “cancer villages” along the Huaihe River.

China’s economy illustrates the problem of growth measured in numbers versus measured in real economic change. The surge in fixed asset investment carried out post-global crisis resulted in an inflation of growth figures, despite the creation of uninhabited apartment buildings, or even entire cities. This is socially unproductive growth, wasteful production, “bad” or false growth. Although the distinction between “good” and “bad” growth exists only in theory, it is essential to clarify the difference to the public in order to move along the path of long-term development.

Admittedly, it may be overambitious to request that a more comprehensive view of growth penetrate the media. However, it would benefit our understanding of China’s economic performance; reconceiving growth would increase competition to generate “good” growth and discourage the race to build businesses that produce “bad” growth.

Yes, I agree. It is indeed an ambitious project, the idea that we could create more comprehensive measures of growth that help us to separate healthy improvement in well-being from cancerous, destructive economic growth. But it seems to me no more an ambitious goal than the vision less than 100 years ago to create the national accounts systems and begin collecting the data (from whence we get GDP measures).

Not Performing Up To Potential

When I was kid there was a comment I dreaded but got too often on too many grade reports to my parents: Not performing up to potential. I hated that. I must say, though, that there times when it did motivate me to do better.

The same comment, not performing up to potential, can easily be applied to the U.S. economy for at least the last 6 years.  I really wish it would motivate our economic policymakers to do better, but alas, they seem to be indifferent to the challenges.

For the details of just how much we are under-performing, I give you the Center for Budget Policy Priorities summation of the latest Congressional Budget Agency report on the economy (below the fold): Continue reading

Busting the Medicare Myths – Presentation

I gave a presentation today to the Michigan Intergenerational Network at Madonna University on the economic prospects of Medicare (U.S.). Thanks to the Madonna Univ. Gerontology Department for support and assistance.

For a downloadable and viewable copy of the presentation, see:  https://jimluke.com/course-resources/presentations/busting-myths-about-medicare/.

Yes, Inflation/Deflation is Hard to Measure

One of the hardest concepts for Principles students, politicians and pundits, oh heck, just about everyone to fully grasp is inflation.  A big part of the reason is because inflation is an abstract concept that is not directly measurable.  We can conceive of it, but we can’t measure it.  I’m no physicist (and open to correction) but it strikes me that it’s a bit on par with “momentum” or “latent energy” in physics.   We don’t have direct-measuring energy-o-meters.  We measure the effects and infer the energy.  Inflation is similar.  We can conceive of a generalized, across-the-economy, sustained trend pushing all/most prices upward such that the unit of money is losing real value in general terms.  Inflation is the sustained push behind all prices. We can’t measure that directly. But we can measure the effect it has: rising prices. The problem comes in that not all prices will be rising at the same time or by the same amount.  Further, during any time period, at least part of the change in price for any good is it’s change in real price relative to all other goods (supply and demand as taught in micro).

We try to deal with this measurement issue by creating a price index – an index that tracks the changes in shopping list of goods over time.  But any price index is a just a subset of all the prices.  Even the Billion Price Project index at MIT admittedly misses most services and lots of consumer goods that aren’t available online.  Price indices are very imperfect beasts.  They have many faults, not the least of them being that they often tend to be volatile in nature.  Since we’re looking for an estimate of inflation which means sustained increases, we need to massage the data further by creating some kind of “core inflation” measure or “trimmed means” type price index.  I’ll explain those some other time.

What prompted today’s post is an article in Bloomberg and a post by Krugman about it.  Together they illustrate one of the reasons so many people want to believe we have greater inflation than we really do.  Companies like to disguise price changes.  They don’t want to be known that prices could be cut in response to demand. Example: auto company offers $2000 rebate on $20,000 car but won’t cut price by 10%, or a firm offers a “value meal”, or they offer a freebie bundled product.  Similarly they often disguise price increases by reducing sizes or portions or by changing the financing.  From Krugman:

Good article in Bloomberg:

Procter & Gamble Co.’s failure to raise the price of Cascade dishwashing soap shows why investors are buying Treasuries at the lowest yields in history, giving the Federal Reserve more scope to boost the economy.

The world’s largest consumer-products company rolled back prices after an 8 percent increase lost the firm 7 percentage points of market share. Kimberly-Clark Corp. (KMB) started offering coupons on Huggies after resistance to the diapers’ cost. Darden Restaurants Inc. (DRI) raised prices at less than the inflation rate as patrons order more of Olive Garden’s discounted stuffed rigatoni than it anticipated.

This is basic economics; prices tend to fall, or at least slow their rise, when there is vast excess capacity and weak demand.

As both the article and Krugman’s excerpt show, we’re closer to deflation than most people realize.  They don’t see the failed attempts to raise prices.  They don’t see the shifts in portions or increase in coupons that reduce effective prices.  What they do see and remember is the $.50 increase in a loaf of bread or the $.70 increase in a gallon of gas.  But even with the gas, they selectively remember the $.70 price increase in summer, but forget the $.75 price drop in autumn.  Inflation and deflation are tricky things to measure.

 

Government and the Slow Jobs “Recovery”

Government finally starts to get out of the way of recovery. In an earlier post today on the good news of the January 2012 employment report, I observed that one of the major factors resulting in an improved (but not good enough) jobs report was that government employment numbers stopped dragging down the total.  I wanted to briefly expand on that idea here.

First, let’s make no mistake the “recovery” from this last recession has been very, very weak.  Private sector growth has been anemic at best. In employment, the recovery has largely been missing in action.  Today, 31 months after the supposed end of the recession, we have only recovered 1/3 of the jobs we lost during the 19 months of recession. As I’ve mentioned before, we are well on our way to a lost decade or more before we regain full employment.  A huge part of the weak recovery has been slow and at times negative growth in private sector employment.

But a bigger problem has been government.  Government has a three-fold impact on employment during a recovery.  Government spending by itself will create employment in the private sector.  For example, if the government chooses to react to a recession and high cyclical unemployment by increasing it’s spending it can create new private sector employment. This would be a classical stimulus program.  The government could embark on highway, bridge, or school construction.  The spending with construction contractors causes those contractors to hire employees. That’s direct private sector employment through government spending.  As long as there are significant unemployed resources (workers), such government spending will increase employment.  Arguments about crowding out do not apply when large unemployed resources exist.

The increased government spending then has a second effect, a “multiplier” effect.  The multiplier effect reflects the idea that workers who got jobs in the initial round of spending themselves spend their incomes and create more demand for more goods. This increased demand for goods results in even more employment.  In other words, the construction workers hired to build the new bridges or schools spend their paychecks.  The firms selling those workers goods then have to hire in order to produce the goods/services the construction workers want.  The exact size of the multiplier effect is uncertain and subject to dispute depending on the econometric methods used to measure it.  However, it’s clear that as long there were substantial unemployed resources to begin with, there is a positive multiplier effect on private employment from increased government spending.

But what I want to draw attention to today is direct employment effect of government.  One of the greatest reasons why we have had a very slow employment recovery is because government in the U.S. has been aggressively cutting jobs for the last 2-3 years. Conservative critics of government have been partially right. Government has been part of the problem – but not in the way they think.  Let’s look at total government employment in recent years:

The data series can be a bit tough to read because government employment has a very seasonal pattern to it.  That’s shows up by the regular up-and-down pattern each year.  Let’s focus on the trend, smoothing out the ups-and-downs. There’s four patterns. Government employment was essentially flat in 2002 and 2003.  Then a period of employment growth in government began running form 2004 through early 2008.  During the recession itself government employment was essentially flat.  Since 2009, though, government employment has been declining.  Cutting government employment is contractionary.  It directly reduces retail demand for goods and services by reducing the incomes of what were formerly government workers.

The pattern is a little clearer if we look at the data in a slightly different way.  The following graph, courtesy of Menzie Chinn at Econbrowser.com, shows the a smoothed trend.  It does this by plotting the 12-month change in government employment (000’s of jobs) by month.

While private employment continues to grow, government employment continues to fall; the decline is most pronounced at the state and local level (Wisconsin is a good example of the contractionary impact of such measures [1] [2]). However, civilian Federal government employment is also declining.

janempsit3.gif
Figure 3: Twelve month change in government local employment (blue), in state employment (red), and government employment ex.-temporary Census workers (geen), 000’s, seasonally adjusted. NBER defined recession dates shaded gray. Source: BLS via FRED, NBER and 

One thing I particularly like about this graph is that it shows the relative contribution of federal, state, and local governments. What this graph shows is that before the recession (the grey zone), government was net hiring approximately 250,000 additional jobs per year. Of that, most was at the local level and some at the state. Very little was federal hiring.

Since the end of the recession in June 2009, government has been firing more workers than it hires.  It has been reducing employment.  The federal government, contrary to popular belief, began shrinking (in employment terms).  State governments were largely able to hold the line on employment until early 2011.  Then state governments began reducing employment in rapidly increasing numbers.  But the big impact again came from local governments.  For the last 30 months, they have been laying off large numbers of workers. The reductions have slowed in 2011, but they are still cutting workers at nearly the same rate that they added them in 2007 – hundreds of thousands of lost jobs each year.

There is a temptation among politicians and commenters to think of government employees as representing largely just some bureaucrats mindlessly pushing paper in large bland office buildings.  That is not true.  At the federal level, most federal government employees are either soldiers or part of some security forces (TSA, FBI, ICE, etc).  At the local level, the vast majority of local government employees are police, fire and emergency workers, and teachers. Reductions in local government employment directly translate into fewer services and less education for children.

Why are state and local governments cutting employment?  Simple.  It’s reduced taxes combined with balanced budget requirements.  State and local governments, unlike a sovereign national government, must balance their budgets.  They are budget constrained.  The recession and weak recovery have hit income and sales taxes hard.  Even more significant is that the collapse of home prices a few years ago has translated into lower property tax collections.  Either way, state and local governments have been pinched.  The response has been to reduce government employment – fire police, firefighters, and teachers.

Paul Krugman notes the how this reduction in state and local government revenue has translated into reduced spending, which in turn has translated into lower employment.  Despite the federal government embarking on a stimulus spending program in early 2009, a program which is over and done with now, it was not large enough to offset the reduction in state and local spending.

if you look at what’s being cut, it’s heavily focused on investment:

That is, we’re sacrificing the future as well as the present. Oh, and the cuts that aren’t falling on investment in physical capital are largely falling on human capital, that is, education.

It’s hard to overstate just how wrong all this is. We have a situation in which resources are sitting idle looking for uses — massive unemployment of workers, especially construction workers, capital so bereft of good investment opportunities that it’s available to the federal government at negative real interest rates. Never mind multipliers and all that (although they exist too); this is a time when government investment should be pushed very hard. Instead, it’s being slashed.

What an utter disaster.

On this point, I have to agree with Paul.  Unless we reverse course and do it strongly, we are flirting with a long-term disaster.  We are under-investing in our future.