Home > Uncategorized > Which base year?

Which base year?

from Blair Fix, Jonathan Nitzan and Shimshon Bichler   

But there is a slight conceptual problem. It turns out that the growth of real GDP – ostensibly a single, objective quantity – is highly sensitive to our choice of base year.

To illustrate, consider a hypothetical economy that produces only two commodities: 1,000 lb of tomatoes and two laptops. Next, let’s choose 1990 as our base year and assume that tomatoes in that year cost $2/lb while a laptop costs $2,000. In this case, real GDP, denominated in 1990 dollars, would be $6,000 (=1,000 x $2 + 2 x $2,000). Now, skip to 1991 and imagine that, in that year, the economy grows by producing one additional laptop. This increase means that real GDP in 1991, denominated in 1990 prices, is $8,000 (=1,000 x $2 + 3 x $2,000). Compared to 1990, real GDP grew by 33.3 per cent.

So far so good. Now, instead of using 1990 as our base year, let’s use 1991. Production levels remain unchanged: 1,000 tomatoes and 2 laptops in 1990, and 1,000 tomatoes and 3 laptops in 1991. Base-year prices, though, are no longer the same: in 1991, our newly chosen base year, tomato prices double to $4/lb, while laptop prices are halved to $1,000. Under these new conditions, real GDP for 1990, this time denominated in 1991 dollars, is $6,000 (=1,000 x $4 + 2 x $1,000), while real GDP for 1991, also in 1991 dollars, is $7,000 (=1,000 x $4 + 3 x $1,000). Unlike before, in this example real growth is only 17 per cent.

And this is the simplest of examples. A slightly more involved example – for instance, one in which the production of laptops is rising and of tomatoes falling – might yield positive real GDP growth with one base year and negative real GDP growth with another.

In other words, real GDP is affected not only by the actual quantities being produced, but also by our choice of base year. And since there are numerous base years to choose from, the same real GDP can end up having many different magnitudes!


  1. lobdillj
    July 28, 2019 at 10:01 pm

    Another alarm bell–We assume, as we should, that GDP factors are ONLY those that are associated with the REAL economy, because the RENTIER economy is parasitical on the REAL economy. The trouble is, official GDP INCLUDES the activities associated with the Wall Street casino, so when the US gifted Wall Street crooks with about $26T in bail outs associated with the great crash of 2008-2009 that was considered income in GDP calculations.

  2. July 29, 2019 at 12:08 am

    For a detailed elaboration of this basic point, see my post on “Real Statistics (3/4) Statistics as Rhetoric” – I argue that most numbers we use in economic analysis are number fictions, not number facts. They are fictions because they incorporate many arbitrary judgements which represent hidden values. However, these fictional numbers are presented as objective measurements of an external reality about which there is no doubt or disagreement possible. The rhetorical strategy of conventional statistics (which is nominal, instead of real) is to HIDE subjective judgments and pretend that all numbers, and all statistical analysis is objective, instead of subjective.

  3. Ikonoclast
    July 29, 2019 at 12:11 am

    Jonathan Nitzan, Shimshon Bichler et. al. have certainly convinced me of the validity of their analysis. It’s hard to see how any person following the precepts of science and empiricism, and having an understanding of the difference between the real and the notional could not be convinced by their work on this matter (GDP) and also by their general solution to the “value controversy” (related to “the capital controversy” or “the two Cambridges debate”).

    It’s correct, as lobdillj says, that official GDP includes those activities associated with the Wall Street casino and the Rentier economy. More broadly it includes the whole FIRE (Finance, Insurance, Real Estate) sector. And we see, for example, that differential inflation as differing inflation rates in different sectors over time (another topic Nitzan, Bichler et. al. deal with) has further compounding effects. Real Estate bubbles are often not counted in CPI figures (due to CPI rigging) and yet that inflation still feeds through to incomes (real estate agent fees and profits on R.E. transactions). Thus GDP is boosted by these effects too and yet the CPI is under-measured.

    Following Nitzan and Bichler’s emphases and analyses on differential profit and differential inflation, we can note that differential rigging of macroeconomic indicators is also embedded in the current conventional system of reporting national economic indicators. In terms of official figures, GDP is systemically rigged high (that is the very reasonable suspicion) by chained base year manipulations and CPI is systematically rigged low by hedonic measure manipulations and selective inclusion or exclusion of goods and services in the official and headline inflation measures.

    Then of course we know that all sorts of negative externality costs as real human costs and real environmental costs are not measured correctly. Indeed, where they are measured at all as in personal, social and environmental amelioration, recovery and remediation costs they also get added into the GDP!

    It is clear that our macroeconomic measures are failing comprehensively and not at all telling us what is really happening in our nations and on this planet. What is really happening is unknown except where people have done empirical studies. Examples are Piketty’s work on increasing inequality and all the scientific work done on climate change, environmental destruction, species extinctions etc. All of this empirical work points to the social and ecological unsustainability of the current global economic system no matter what the concocted formal, notional numbers of the dollar numeraire say.

    In summery, the macro numbers of conventional economics say we are doing great, at least if you don’t drill down into the layers of inequality. The empirical data from science (and from empirical political economists) show we are approaching a catastrophic asymptote.

  4. Ken Zimmerman
    August 2, 2019 at 12:17 pm

    Lobdillj and Ikonoclast, if this kind of thing bothers you, you might consider just measuring how much of each product is created each year. But that does not cure the problem of estimation and inexactness in mathematics-in-use. The base year chosen is just one such issue with GDP. Others, related to estimation-inexactness include: is the data fudged, did anyone lie about what was produced, does the product have to be usable to count, what is the source of pricing, etc. The same issues arise with the simpler process of counting each product per year. Plus, a few more issues. How are changes in the price of a product during the year handled and how to price products that are given away (charity) or stolen. These demonstrate the malleability of mathematics. As jobs in which math is used change, so mathematics may change. Also, mathematics is subject to hierarchical controls, like other aspects of our lives. If a supervisor instructs one not to utilize a formula or data set, is one to dissent? Even the queen of all the sciences, physics must deal with all these issues. For example, check out the TV drama “Manhattan.” Two things are different about this drama from other dramatic treatments of the Manhattan Project. First, the physicists, military, FBI, etc. involved in the project are three-dimensional. Each is involved in many actions and judgments apart from physics. And it shows the physics work as three-dimensional, as well. Some of the math is accepted by all, or most of the physicists involved. Some is not. Plus, physicists emphasize different aspects of any accepted formula, depending on what each believes is important, and what each wants to achieve. It is Robert Oppenheimer who channels these different perspectives. Not an unusual role for a person placed in charge of a group of talented and aggressive high achievers. The dynamics of this society are complex and always on the edge of collapsing. In fact, if it was my intent to lie about something important, I can think of no better tool to achieve that goal than mathematics.

    Asad, as to number facts vs. number fictions, you’ll need to explain to me how the two are distinguished from one situation to another, and one time to another.

    Overall this story expresses my views on mathematics. In ‘The Presentation of Self in Everyday Life’ US sociologist Erving Goffman includes a chapter called “Regions and Region Behavior.” There Goffman develops the concepts of ‘front’ and ‘back.’ In a restaurant the dining area is the front; the kitchen is the back. In a theater, front stage is for the audience; backstage is for actors and stagehands. In front, actors (waiters) wear costumes (uniforms); in back, they change clothes or wear casual dress. In front the public is served; in back, professionals prepare to serve them. Front is where the public is admitted; back is where it’s excluded. Goffman’s contribution is extending “front” and “back” from restaurants and theaters to all our institutions. In the university, classrooms and libraries are the front, where the public (students) is served. The chairman’s and dean’s offices arc the back, where the products (classes and courses) are prepared. This separation is a necessity. Goffman gives an example of distress from mixing front and back: a gas station where customers walk into the toolroom and help themselves to wrenches. He quotes George Orwell: “It is an instructive sight to see a waiter going into a hotel dining room. As he passes the door a sudden change comes over him. The set of his shoulders alters; all the dirt and hurry and irritation have dropped off in an instant. He glides over the carpet, with a solemn, priest-like air … he entered the dining room and sailed across it, dish in hand, graceful as a swan” (p. 121). The purpose of separating front from back isn’t only to keep customers from interfering with the cooking. It’s also to keep them from knowing too much about the cooking. Everybody down front knows the leading lady wears powder and blusher. They don’t know exactly how she looks without them. Diners know what’s supposed to go into the ragout. They don’t know for sure what does go into it. Traditional philosophy recognizes only the front of mathematics. But it’s impossible to understand the front while ignoring the back. The front and back of mathematics aren’t physical locations like dining room and kitchen. They’re its public and private aspects. The front is open to outsiders; the back is restricted to insiders. The front is mathematics in finished form-lectures, textbooks, journals. The back is mathematics among working mathematicians, told in offices or at cafe tables. The front divides into subregions-first, second, third class. A restaurant may include banquet hall and snack bar. Theaters can have box, orchestra, and balcony. cony. The mathematical public includes professionals, graduate students, and undergraduates. The back also divides into subregions. In a restaurant, there are domains of salad chef, pastry chef, and dishwasher. Mathematicians divide into finite group theorists, numerical linear algebraists, nonstandard differential topologists, and so on and on. Front mathematics is formal, precise, ordered, and abstract. It’s broken into definitions, theorems, and remarks. Every question either is answered or is labeled: “open question.” At the beginning of each chapter, a goal is stated. At the end of the chapter, it’s attained. Mathematics in back is fragmentary, informal, intuitive, tentative. We try this or that. We say ‘maybe,’ or ‘it looks like.’ This distinction is described by George Polya in his preface to Mathematics and Plausible Reasoning. “Finished mathematics presented in a finished form appears as purely demonstrative, consisting of proofs only. Yet mathematics in the making resembles any other human knowledge in the making. You have to guess a mathematical theorem before you prove it; you have to have the idea of the proof before you carry through the details. You have to combine observations and follow analogies; you have to try and try again.” Mainstream philosophy doesn’t know that mathematics has a back. Finished, published mathematics–the front–is taken as a self-subsistent entity. Worse, mainstream philosophy doesn’t look at the real front–real articles and treatises. It contemplates idealized texts as logicians would have them, not mathematical texts as they are and must be. It would make as much sense for a restaurant critic not to know there are kitchens, or a theater critic not to know there is backstage.

  5. Blissex
    August 3, 2019 at 7:40 pm

    The base year is the least of the problems with the GDP index, the biggest problem is that the GDP index is actually mostly the GDI (Gross Domestic Income) index, that is it does not track production, but income.
    The confusion between the is based on the neoclassical assumption that the domestic economy is made of perfectly competitive trading markets, and therefore the purchases of every market participant are also the income of some other market participants.

    GDP is more properly the list of physical quantities (e.g. tons of plastic, hours of surgery, km-ton of freight, …) produced by a political economy. Every attempt to turn it into a monetary index, by multiplying it with some arbitrary price vector, or replacing it with aggregate income, severely reduces its usefulness, which is somewhat limited.

    JM Keynes in chapter 4 of “The general theory …” explicitly rejects both GDP, its various indices, and GDI as useful for moderling purposes, and proposes to use (weighted) hours of work instead.

  6. Blissex
    August 3, 2019 at 7:47 pm

    There are some interesting physical proxies for actual GDP and one of them is the physical quantity of electricity consumed, and in particular the quantity per person; the reason is that electricity consumption is subject to Jevon’s Effect, which sort-of says that some things are so useful that demand for them expands as much as possible.

    The World Bank uses electrcity consumption per person as a development indicator, and Google have published a nice interactive form for that, and there is in that something deeply worrying: for the first time in recorded history electricty consumption per head has dropped drastically below trend in many “first world” economies, and obviously that is not because of more efficient technology, compare the lines in this graph:


    • Ken Zimmerman
      August 4, 2019 at 2:55 am

      Blissex, the decrease in energy use per person is in part a result of energy efficiency use improvements in technology in everyday use. But it’s complex. It involves not just energy equipment efficiency improvements but also the number of people using such technology and how they use it on a daily basis. It’s a cultural thing. Electricity use is the focus of much of this research.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.