Economists Do It With Models

Warning: “graphic” content…

Bookmark and Share

October 25th, 2015 · 2 Comments
Macroeconomics · Tumblr

http://economistsdoitwithmodels.tumblr.com/post/131906571950/the-57-year-old-chart-that-is-dividing-the-fed

Tags: Macroeconomics · Tumblr

2 responses so far ↓

  • 1 Jeff // Oct 25, 2015 at 8:42 pm

    I just looked at the attached website. It has no comments area. Obnoxious – how do you point out errors? I guess that Aggregate demand will include ” and not exports (total exports minus total imports)” for the foreseeable future, then… Or, more to the point, I really wish that he defined Aggregate demand as including “investment in physical capital” instead of just investment. I hope that economists and non-economists can agree that buying Intel at 100 isn’t aggregate demand.

    The (A?) problem with the Phillip curve is obvious if you look with just the right perspective. Imagine that “natural” unemployment is at 4% and current unemployment is at 5%. I am ignoring the fact that unemployment is not exactly defined here.

    There are at least two different ways to get from 5% to 4%. One approach is for an increase in demand to trigger an increase in hiring. Another approach is for an increase in desperation among workers to trigger a decrease in the level of wages one consider’s acceptable (the Nike sweatshop model).

    Mathematically, these cannot be differentiated based on just the info in the original Phillips curve equation, so the degree to which the Phillips curve has a chance of being right is upper-bounded by the degree to which the former is the path to lower unemployment and not the latter.

    When seen in this light, statements like “What economists initially failed to realize in constructing the Phillips curve was that people and firms take the expected level of inflation into account when deciding how much to produce and how much to consume. ” are less than helpful.

    If you truly want to predict consumer price inflation (I hate the undecorated term inflation – give a word too many meanings and it has no meaning at all), then cut out the inadequate middle man of unemployment and go all the way back to Aggregate Supply and Aggregate demand.

    So, Aggregate Supply is supposed to be equal to aggregate demand. How does this help? We remember from inorganic chemistry that any reaction is driven by its rarest ingredient.

    The consumer part of aggregate demand consists of the inner product of consumer preferences (a near constant) with individual wages plus capital gains available to the middle class (not in the least bit constant. Aggregate supply, of course, is driven by the inner product of physical investment opportunities with the expected consumer demand against those opportunities. And I do mean physical capital opportunities and not financial capital opportunities. The presence of a need for two Taco bells can be the difference between just building two restaurants (as happened before the Reagan Revolution), and bidding one restaurant up to twice the price (as happened after the Reagan Revolution).

    So… getting back on track. How can this be used to predict inflation? Consumer price inflation (as opposed to asset price inflation as referenced above) happens when the factors of production (physical capital and the wages utilizing that capital) are tight while consumer budgets are loose.

    Note that this can happen in both high and low unemployment scenarios! If labor is the limiting factor for production, then indeed tight labor can be inflationary, especially if it drives higher wages at the same time. But if physical capital is the limiting factor for production, then it is possible that rather than driving inflation, tight labor could drive economic growth instead. (Is there a good reason for the Fed to limit that?)

    Conversely, tight consumer budgets, which can happen in both low unemployment and high unemployment environments (consider that the job market for an individual is within an industry; not within a country), can invariably be counted on to drive low inflation levels.

    A much more useful predictor of inflation might be the ratio of middle class savings (which, at least since 1980 has varied inversely with the average savings rate for the country) to underutilized capacity of physical capital and real estate.

    Given that tightening budgets and greater desperation can have the same effect as loosening budgets associated with greater employer need, it really does make sense to cut out the middle man in this case.

  • 2 Jeff // Oct 25, 2015 at 9:47 pm

    I apologize. I thought that the quote above on businesses and consumers incorporating inflation expectations into their production/consumption decisions was from the other poster (MM, who did some of the definitions). If I had been aware of the author I would have been more constructive in my criticism. Anyway, to defend myself, …

    Incorporating inflation into a decision to go buy a pint of Haagen-daaz clearly isn’t meaningful, but you could potentially model the quote above looking at decisions like those to make purchases that will create physical capital. Taking that perspective, what would said incorporation look like?

    If cost of goods sold go up and revenue goes up, then these would cancel out, but you would have to add an extra cost adjustment – that is the delta in your cost of supplies between when you buy them and when you sell the finished goods. Since you make extra holding the goods during inflation, you would expect compensation for this to be your supplier saying “You need to pay me more now because I expect you will make more later”.

    Ok. Fair enough. If anticipated, inflation moves through the system faster.

    Another way of looking at this is that supply and demand curves will seek equilibrium based on future prices rather than today’s prices.

    So… There is some meaning to the suggestion that the system will take into account the inflation rate.

    However, looking at things from another perspective, unemployment above frictional levels can be thought of as by definition an indication that supply and demand are not in balance. After all, the value of the unemployed person’s labor has to be measured as greater than zero.

    The key question is not “What is the equilibrium target”, but rather “Are we moving toward (lower unemployment) or away (higher unemployment) from that target?” This of course assumes that “equilibrium target” is a well-defined concept, but that’s a journey for another day.

    So, what is the connection between taking expected inflation into account and unemployment levels?

    The key insight here is that with a well-timed switch to micro, each business has a change in cost levels associated with holding time of their assets and their inflation level that can be assumed to be a “constant” in the same sence that a mass that undergoes relativistic accelleration will have a “constant” new mass. Accellerate back to the original levels or switch your inflation rate back to original levels and this change will be equally and oppositely adjusted. With a new cost structure, the company will have a new ideal set of employment levels. However this “ideal cost structure” is completely orthoganal to the underlying purposes of loose vs tight money – that is, to impact the creation of new business and alter the rate at which marginal businesses are driven into bankruptcy. That means that the constant part of the equation – the equilibrium part that is not really impacted by tight or loose money – is exactly the same as the part that adjusts to the inflation rate. The original input of money being tight or loose has not been cancelled out in the process.

Leave a Comment