John Irons's Blog


Economic News, Data and Analysis


The US Labor Department Thursday announced that the revised third quarter
productivity growth (SAAR) was 4.5%. This is very good
for the economy. While news reports do a good job of announcing
the latest numbers, they often fail to provide a historical context in
which to evaluate the information.

Below is a graph of US non-farm productivity going back to 1948. As
you can see, productivity growth was strong just after WWII and remained
relatively strong until the early 70’s when it began to dip. We’ve seen
growth at the current level for only two quarters in the past ten years.

Hopefully the strength in productivity growth (2.5% over the past year,
and 4.5% in the third quarter) will continue to build.

Source: Quarterly
Labor Productivity

Where to go for Productivity Data?

The productivity data for the US can be found at the Quarterly
Labor Productivity Home Page
from the Department of Labor. For data
from other countries take a look at the OECD
web site for many of the basic series including the national
and related data. Stat-USA
has data from many government data series on-line for registered subscribers
and the NIPA Visualization
also has a number of series as well.

Other parts of this site have more resources on economic
and government agencies.


This is the point where most commentators take a leap into the unknown
by predicting how the economy will do in the future. Some say that good
times are ahead in the “new economy” as new technologies will drive us
to a new era of growth. Some predict gloom and doom or claim recession
lies ahead because of global competition, runaway entitlements, or even
the year 2000

I won’t bother to leave myself open for error, at least not today.

When all is said and done, productivity growth is perhaps the most important
gauge of the performance of the economy. Let’s hope that the strong growth

See Also:


Productivity Growth Matters…
a lot. 

Number of years to double your standard of living: 


Productivity Growth Time to double Economic Standard of Living




139 years 

47 years 

29 years 

21 years 

16 years


Economic output in 50 years: 

(relative to today with same labor input, Growth: 0.5% – 4.5%) 


Filed under: Data

The Economics of On-line Publishing

Don’t expect free books anytime soon.

It’s the year 2025. Politicians are grappling with how to deal with
a social security system that is threatening to eat 50 cents of every dollar
earned by the working population. “Books” are no longer printed on paper,
but are rather stored electronically on the new “Octium IV” computer sitting
in your shirt pocket. Each page of the book is transmitted from your pocket
to the ultra-thin hi-resolution folding screen on your lap so that you
can read your favorite author on the Metro on the way home from work.

By many accounts, we are headed into a time in which on-line publishing
and “electronic books” will come to dominate the more traditional forms
of print communication. Cloth books on your shelves and magazine on your
tables are thought to be nearing obsolescence. As we move into this new
publishing world, what will happen to the price of new books and how will
standard economics cope with the new characteristics of the industry?

It has often been claimed that we are entering an era in which the physical
costs of a new book — such as transportation, printing, paper and ink
costs, and warehousing overhead from storing books — will drop to near
zero. (The internet cuts down on shipping costs and once a book is electronic,
it can be stored, and “published”  for essentially nothing.) 
As such, it is thought that standard economics does not apply, since economic
theory often operates in a world of “increasing marginal costs” and some
claim it is ill suited to deal with a product in which the very first copy
entails hugh costs, while each copy thereafter is essentially free to “produce”.

Upon closer examination, there appear to be two problems with this story
line. First, this brave new world of publishing looks very similar to the
one we have now, and second, the basic economics
of the industry become easier, not harder. When the marginal cost of a
product drops to zero, we only need to look at the demand side to determine
prices. And once the demand side is taken into account, one would not expect
the price of books to merrily follow the production costs down to zero.

Not much will change.

The cost of producing the copy of the book on you lap on the metro in
2025 is just about zero. The MIT
, who is the publisher of this title, was able to upload the book
to your computer in a second or two without any human interaction. The
real cost to the publisher came from paying the author, doing the marketing,
editing, etc. All these costs came before the first copy was sent to the
reader – and no real costs are occurred after that in getting the book
to you.

On the surface, this may seems like a completely different set of rules
for an industry, and we should expect major changes as the transition from
paper to silicon takes place.  But is this really much different from
what going on now?

Last spring I bought a gift for my younger brother’s college graduation.
It was about 200 pages long, had high-quality paper, a nice design on the
cover and was hard bound of course — nothing but the best for my siblings!
The unique thing about this book was that each and every page was completely
blank – it was a journal for him to record thoughts, and ideas in his spare
time. And it only cost me 8$! (and, yes, I got him something else too).

Now, looking at my bookshelf,
I can find many books which cost 5 times that much – and several that cost
10-15 times that for something of about the same quality in terms of the
physical aspects of the book. What must therefore be driving up the price
that I pay for a good read is the content provided by the author.
Say that I was able to magically transfer all of my books to electronic
from and redeem the 8$ per copy that I spent on paper and cover. Although
I would like the extra cash – I would also find that I had still spent
a small fortune on the content of the books alone, and independent of the
physical costs of the book.

Whether the book’s materials and shipping cost the current 8$ (plus
a small amount to put the ink on the page) or 4$, or 0$ doesn’t really
matter for either the publishers (who get many times that for selling a
book) or the reader (who is willing to pay many times that for the right
to read it).

Also, if we could magically eliminate all the physical costs of producing
a book today, and all those cost reductions are transferred
instantly to the readers
, we would still see prices at nearly the same
level. Production costs are simply a small fraction of the price we pay
for books.

So why do we pay so much for a book?

Books are a monopoly.

In many industries the products are all the same or very similar. The
company who offers me the product at the lowest price will get my business.
If a company is charging more that it costs to make the product – other
companies will be able to steal their business, and in the process drive
down prices.

Publishers, however, are able to offer a unique product to their customers
– the work of an author. In economic terms this means that the seller faces
a downward sloping demand curve. This allows the publisher to charge more
for a book than the costs needed to produce it without fear of another
company under-cutting its prices – since another company can’t offer the
same product.

Since publishers are out to make a buck, they will be able to successfully
keep the price above it’s costs. In this case the price and the profits
are primarily determined by the demand for the book – and not the
cost of the supply of the book. [This idea can be formalized
using basic economics if you know a tiny bit of calculus].

Thus, reducing the cost of each additional book will not necessarily
have a major impact on the price
. The error in the “free books” school
of thought is ignoring that all books are not substitutes and the owning
the right to publish a particular book means that the price of that book
can be higher than the costs of production.

What will change…

Publishers now hold power over authors, in that good deal of up-front
start-up cash is required to get a book initially printed. As these costs
drop, the authors will be able to use the threat of self publishing more
effectively in bargaining. Publishing houses will then have to become more
like recommenders and advertisers in their business than as a company that
specializes in putting ink on paper.

As the initial outlay for publishing drops, we can expect more “publishers”
and nontraditional publishers and self- publishers getting into the game
by offering services in the recommendation and promotion aspect. This may
eventually lead to more selections and a wider range of books brought to
the market. It may also lead to better paid authors since they will have
a stronger bargaining position with the publishers.

However, don’t expect terrifically cheaper books anytime soon.

See Also:


Formal Model of profit
maximizing publishers


Assume that the publisher wishes to maximize profits for some book.
What will happen to the price of the book as costs drop to zero? 

Total revenue for the book sales is the price times the quantity sold
(which depends on the price of the book): P*Q(P). Q(P) is the demand curve
for the book in question. Total costs depend on the total number of books
sold. So profit = revenue – costs, or 

(1)  Profit(P) = P*Q(P) – C(Q(P)) 

maximizing profits mean setting the first derivative to zero: 

(2)  d Profit / d P = Q(P) + Q'(P)*P – C'(Q(P))*Q'(P) = 0 

Which solves for the optimal price P. 


(3) P = – Q(P) / Q'(P) + C'(Q(P)) 


Now consider the case where each extra book costs nothing to produce.
This new price will be called p Equation (1) becomes 

(1′)  Profit(p) = PP*Q(p) – 0 

(2′)  dProfit / d p = Q(p) + Q'(p)*p = 0 


(3′)  p = – Q(p) / Q'(p) 

First note that since the cost of producing an additional book today
is positive, C’>0, and we have P>p. So the price of books will drop. 

If we further assume that the demand curve is isoelastic (i.e. Q'(P)
/ Q(P) = constant ). Then  we know that the difference in price is
P-p = C'(Q(P)) and the price drops by exactly the same about at the marginal
unit cost of the book. 

In general, the price drop will depend upon the shape of the demand

Exercise: What must be true of the demand curve to be able to
get p arbitrarily close to zero?

Filed under: Microeconomics