Discussion Paper

No. 2012-13 | February 14, 2012
A Case Where Barro Expectations Are Not Rational

Abstract

The paper generalizes Feldstein’s criticism (Perceived Wealth in Bonds and Social Security, 1976) of Barro’s analysis (Are Government Bonds Real Net Wealth?, 1974) for the case that the interest rate exceeds the growth rate. This is done by considering an economy in steady state where all agents hold 'Barro expectations': they believe that government debt must necessarily be repaid and therefore leave the present value of their income streams unchanged. In this scenario, a change in the mode of taxation affects the present value of disposable income in the private sector. This violates their Barro expectations.

See the revised version of the paper in the comment section below.

Data Set

Data sets for articles published in "Economics" are available at Dataverse. Please have a look at our repository.

The data set for this article can be found at: http://hdl.handle.net/1902.1/17615

JEL Classification

E12 E2 E6 H6

Cite As

Ekkehart Schlicht (2012). A Case Where Barro Expectations Are Not Rational. Economics Discussion Papers, No 2012-13, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2012-13

Assessment



Comments and Questions


Ekkehart Schlicht - A generalization
February 19, 2012 - 14:14

I just saw that the argument can be considerably generalized. Assumptions (2) and (3) that postulate exponential growth can be dropped. The result (19) remains valid for arbitrary growth paths.

In view of this observation, I suggest to change the title from "A Case Where Barro Expectations Are ...[more]

... Not Rational" to "Barro Expectations Are Not Rational."


Anonymous - An epistemological aspect
April 11, 2012 - 10:03

The fate of this discussion paper claiming to disprove Barro's
highly influential interpretation of Riccardo-equivalence
could become extremely interesting from an epistemological point of view.
Barro's Theorem can be regarded as an important theoretical rationale for austerity doctrines which
have good chances to be cemented in laws by the European ...[more]

... Union.
Though the issue should be in the focus of economic debates now, nobody seems to take notice of the paper.
It is available now since about 2 months but downloaded less than
100 times and a search for comments delivers exactly comment.


Ekkehart Schlicht - The Pythagoras problem
July 09, 2012 - 08:17

You ask why very few took notice of the paper.

I was surprised, too. My explanation runs like this:

Suppose somebody claims to have disproved Pythagoras' theorem. I would ignore that because I have proved the theorem at high-school and I think I can can write down ...[more]

... the proof.

Apparently some economists consider the Barro-Ricardo equivalence thesis as something akin to Pythagoras' theorem. This would make their disinterest and the intellectual resistance I observe fully comprehensible to me.

Some further thoughts. I sent the paper to many colleagues. Only two of those who replied commented on the subject matter. The majority remained ambivalent about the issue. I think they doubted my example. Of the two comments one was by a colleague from another university who suggested that I should consider a change in the rate of interest as a response to the switch in the fiscal regime. As this would rule out the Barro-Ricardo equivalence directly, I do not think that I shall follow this suggestion, but it made me aware that even quite distinguished economists do not understand the equivalence thesis, but nevertheless accept it. The other comment was by Robert Solow who agreed with my example and added two remarks. This was heartening. As I said, it was the only positive response I have obtained so far.

It appears to me that the validity of the Barro-Ricardo equivalence thesis largely goes unquestioned. I notice that even colleagues who oppose austerity measures do not follow my argument. This is a good sign in the sense that scientific beliefs do not always follow political conviction.

My note is totally destructive and hypothetical. In this sense it does not make any positive contribution and will cease to be of relevance once it is understood that the Barro-Ricardo equivalence does not apply in a growing economy. But the piece may perhaps remain interesting for historians of economic thought. They may want to explain how it could have happen that a relatively elementary mistake went unnoticed for nearly forty years and even became important in policy discussion.

Future historians of economic thought may also try to understand the extraordinary resistance of contemporary economists to accept that the Barro-Ricardo thesis is wrong, and why they, in a majority, stick to an argument that is disproved by a simple example.


Anonymous - Destructive?
July 18, 2012 - 10:22

I do not agree with your statement that this contribution is purely destructive. It generates irritation by challenging obvious 'trivial' knowledge, and irritation seems to be the most important trigger for productive research. Furthermore the current claims for austerity seem to be based on striking logic and evidence. Considering the ...[more]

... destructive potential of such claims it appears to be quite constructive to scratch on their foundations.


Anonymous - Referee Report 1
April 13, 2012 - 08:38

See attached file


Ekkehart Schlicht - Reply to the first referee
April 29, 2012 - 12:29

See attached file


Anonymous - It takes two for tango
July 17, 2012 - 15:58

After reading Schlicht's reply to this reviewer's report, my impression is that he disproved all the criticisms put forward by the reviewer. I am expecting and waiting now for a response of the reviewer.
To the best of my knowledge open assessment journals were established to overcome the monopolistic one-way ...[more]

... communication structures associated with traditional double-blind clandestine reviewing procedures.
My impression from this reviewing process is that the reviewers and the editor should change their behaviour to show that they take the central AIMS and scope of E-COnomics serious.


Anonymous - Referee Report 2
April 13, 2012 - 08:39

See attached file


Ekkehart Schlicht - Reply to the second referee
May 14, 2012 - 09:55

See attached file.


Anonymous - Almos worhtless report
July 17, 2012 - 15:08

After going through the report of this reviewer, I understand why the responsible associated editor hesitates to come to a decision on the discussiong paper. The reviewer first states that Schlicht's algebra is correct. But then he doesn't try to find out whether Schlicht's argument is correct in the sense ...[more]

... that it matches the assumptions of Barro's theorem. He simply restates well known formulas from OLG models. From my point of view,
it is not the task of reviewers to restate again and again what they know. (As he correctly mentions, his formulas could be read elsewhere!). The nature of scientific research and the only source of additional knowledge is to inspect contradictions between new and established theories. Since the current issue is a central one for economics in general, I invite and advise the reviewer to take this chance to DO RESEARCH and to give an answer to the question posed by the editor. If he finds it too difficult to find an answer, it would be quite helpful to admit this. Then someone else could be asked...


Ekkehart Schlicht - A helpful report
July 18, 2012 - 08:08

The report was valuable to me, as it confirmed the correctness of my example and outlined explicitly the reaction I now expect from of almost all economists I spoke to in the meanwhile. This helped me to think about the matter again and look for a way to present the ...[more]

... argument more convincingly.

It is also remarkable that the referee did not absolutely insist that I am wrong, but left this open, in spite of not being convinced by my example. According to my experience, such a self-effacing attitude is quite uncommon for referees.


Ekkehart Schlicht - A quick first reaction
April 15, 2012 - 08:31

I thank the referees for the time and effort they have devoted to critically review my contribution and shall comment on each report in detail later.

Both referees take issue with my formulation of the household's intertemporal budget constraint. I suggested

"The present value of the ...[more]

... households' expenditure is to be equal to the present value of the households disposable income"

and defined disposable income in each period as gross income minus taxes plus income from interest on capital holdings.

Given these definitions, neither of the referee questioned my result, but each remained convinced of the theoretical validity of the Barro-Ricardo equivalence and suggested that I should change the definition of the intertemporal budget constraint. (Both made different suggestions, though.)

This is a valid point. My result depends crucially on the definition of the intertemporal budget constraint chosen. When writing the paper, I was not aware that the definition of the intertemporal budget constraint I used is actually controversial. In any future version I definitely have to clarify this much better, and alert the reader that the relevance of my observation depends on the relevance of that definition, and that the definition chosen is controversial.

I shall therefore enlarge on this issue in any future version of the paper an thank the referees for making me aware of this problem.


Ekkehart Schlicht - Ricardian Expectations
May 31, 2012 - 17:54

The present title has been chosen a little thoughtlessly. My aim was to convey the message clearly, but I have overlooked that the present title may be unfair to Robert Barro in the sense that it contains the notion of "Barro expectations" with a negative connotation. Sorry about this.
As ...[more]

... it was never my intention to criticize Barro, but rather criticise the Barro-Ricardo equivalence thesis, I propose as a title for any revised version

"Ricardian Expectations Are Not Rational."


Ekkehart Schlicht - Revised version
June 30, 2012 - 08:25

Revised version

As I have not heard from the associate editor for quite a while, I have prepared a revised version that takes into account the referees' comments as well as suggestions by other people I have obtained in the meanwhile.

Please find the revision attached.

Here ...[more]

... is the revision report.

1. The most important change is that I have inserted a section that pinpoints the difference between Barro's argument and the result obtained in the example. The reason is that both referees as well as some economists I talked to, were not satisfied with my counter-example and actually continued to believe in the Ricardian equivalence as a theoretical proposition. (Barro's thesis is closely related to the budget constraint the first referee has proposed. I argue that this is mistaken.)

2. I have introduced a section that quotes Barro extensively. This permits me to go through his argument step by step in the context of my example and determine which of his statements are true and which are not. This permits identification of his mistake. It is also of advantage to have a first-hand statement of the thesis which I attack, as this avoids ambiguity regarding the position I critizise.

3. I have changed the term "Barro expectations" to "Ricardian expectations." Maybe the former term is too personal, while David Ricardo is above being identified with the position currently labelled "Ricardian."

4. The notation has been changed slightly, to avoid by some misreadings.

5. The central example remains unchanged.

6. I have added a section that discusses what happens when the tax policy switches back from a debt regime to a pay-as-you-go regime.

7. In view of the importance given by the second referee to the general applicability of Barro's reasoning (with time-dependent interest etc.), I have added a section about the validity of my findings under such quite general conditions. I have adjusted the title to reflect the more general nature of the new statement, and make it less personal.

8. I have not taken up the suggestion of the second referee regarding the budget constraint, as her or his suggestion is equivalent to mine. I have explained that in my answer to the second referee's report.

I note that I have submitted the paper more than half a year ago (February 13th, 2012). Apart from seeing the referee reports (April 13th), I have not heard anything about the submission up to now and would appreciate to learn about the present state of the decision process, or what I can do to help with reaching a decision.


Anonymous - Congratulations
July 17, 2012 - 15:27

The revised version of the paper is really exciting.
The author now cites Barro's argument to pinpoint
the inconsistency in Barro's derivation. The exposition is extremely clear and lucid.
Now it is to be hoped that the associate editor will recognise
the potential of the paper to push the journal ...[more]

... to top ranks and,
more importantly, to advance science and economics a big step forward.


Anonymous - Overstated
July 19, 2012 - 10:55

It should be kept in mind that both the simple growth model anlyzed in the original submission as well as the more complicated Ramsey framework brought up into the discussion by the reviewer lack realism. Considering this, the formulation above ('advance science and economics a big step forward') appears quite ...[more]

... exaggerated. Nevertheless it remains worthwhile to correct mistakes or point to ambiguities in frequently cited contributions as Barro's 1979 paper.


Anonymous - Referee Report on Revised Version
July 18, 2012 - 08:54

see attached file


Anonymous - Missing the point?
July 18, 2012 - 13:50

Here are an ongoer's impressions of this reviewing process:
The submitted paper claimed to show that Barro expectations
are not rational. To my understanding the central question was whether Schlicht's criticism matches and disproves Barro's scenario. The reviewer's answer seemed to be evasive. Instead he emphasized (and now again emphasizes) ...[more]

... that Schlicht's case does not disprove more general notions of Ricardian equivalence in a more general Ramsey model. To my understanding the only things still missing are a definitive answer to the central question and -- depending on this answer -- an advice whether Schlicht should narrow his conclusions or emphasize the narrow scope of his criticism.


Ekkehart Schlicht - Let us be precise and not evade the main issue
July 18, 2012 - 22:29

Thank you.

My original submission did not claim, however, to say anything regarding Barro's analysis, and whether the equivalence thesis is true or not true. I simply considered a standard infinite horizon growth model in steady state and showed that Ricardian behavior entails a violation of Ricardian expectations ...[more]

... within this standatrd set of assumptions, quite regardless of what Barro has assumed or not assumed. I found this interesting and wanted to share it with other economists.

So, in my view, the central question is not, as this commentator puts it, whether my "criticism matches or disproves Barro's scenario". It is simply whether my example is valid or not valid. All discussion so far has confirmed that it is valid.

A next step would be, indeed, to look at "Barro's scenario". This was not the topic of the first version. I actually avoided that question in order not to get involved in a discussion about Barro's writings. I wanted to emphasize the economic substance matter, rather than Barro's way of phrasing things.

Both reviewers pressed me, however, to comment on Barro's mistake (while not identifying my mistake, and there seems to be none). In the revised version I took that suggestion up and referred to Barro's !979 statement. The second referee seems to agree that Barro makes a mistake here. So that is fine with me, and may make it easier for other readers to get into the case. Yet this discussion of Barro 1979 concerns an issue entirely different from the central one outlined above, and distracts from it, just as I feared.

I am still waiting for the (anonymous) editor's response. In order not to induce further delay, I shall not comment on this report on the new version. I note, however, that it does not deal with my example or the generalized example in the new version, but introduces still other assumptions and other notations. This is, again, fine with me and I could comment on that, but it does not relate to the validity or invalidity of my examples.


Ekkehart Schlicht - Reply
July 20, 2012 - 13:08

See the attached file.


Ekkehart Schlicht - Anonymous rejection
July 19, 2012 - 16:40

The paper has been rejected.

I have obtained a rejection mail from the editorial office. As the editor's decision letter has not appeared on this webpage, I publish the rejection text here:

___________________________

Dear Ekkehart:

We are sorry to inform you that based on the ...[more]

... open discussion on our website and the referee report(s) the co-editor in charge has decided to reject your submission, MS 678, entitled

"A Case Where Barro Expectations Are Not Rational"

for publication in Economics.

Please feel free to contact the editorial office if you have any questions about this decision, or if you have difficulty reading any of the attachments.

Thank you for giving us the opportunity to consider your work.

Sincerely,

Korinna Werner-Schwarz
(Administrative Editor)
------------------------------------------
Economics
The Open-Access, Open-Assessment E-Journal

http://www.economics-ejournal.org



Attachments:

R1's report
R2's report

__________________________________________

Included were the two referee reports that are available since April 13 on this page. There was no decision letter. The paper itself has been submitted on January 19, 2012 -- half a year ago.

I have asked the editorial office to reveal the identity of the responsible co-editor, so that I know whom I have address to speed up the evaluation process. This request has not been answered up to now.

Nobody has questioned my example. The (quite competent and helpful) referees proposed changes in my example that could possibly salvage the equivalence thesis. I have shown in my reply that these proposals are not feasible (Referee 1) or equivalent to my formulation (Referee 2)

I have heard not a single word from the responsible associate editor, and certainly not any reason for the decision and the strange delay in taking this decision.

Ekkehart Schlicht


Anonymous - Decision Letter of the Co-editor
July 20, 2012 - 08:03

Based on two referee reports and feedback from one referee on the revised version I have decided not to accept the paper for publication in Economics-Ejournal. Instead I invite interested readers to a further debate of the discussion paper and its implications.
In accordance with the referees and with the ...[more]

... author I see in the centre of this debate the correct formulation of the households intertemporal budget constraint under Ricardian expectations. The referees understand, in accordance with Barro and a bulk of related literature, Ricardian expectations as ruling out that households base their consumption decisions on a disposable income which includes interest income on government debt. The author challenges this view and shows that with the inclusion of interest payments in the households disponsable income the Ricardian equivalence thesis no longer holds. However, this view is based on a re-interpretation of Ricardian expectations which does not seem to meet the original intention of the concept. So it is no wonder that it leads to a different result. Whether the new interpretation and the following results gain broader support within the academic community than the original ones should be left open to further discussions.


Johannes Ludsteck - There is no cat in the empty box
July 24, 2012 - 09:54

If I (a perplexed non-expert) understand the co-editor correctly, he states that the bulk of the related literature bases the analysis of Ricardian expectations on assumptions "[ruling] out that households base their consumption decisions on a disposable income which includes interest income on government debt."
No rational agent will ...[more]

... buy government bonds if he doesn't consider the associated interest payments as disposable income. Then why waste even a single further thought on the analysis of government debt?

Based on this observation I am tempted to conclude that the Ricardian expectations literature went out to demonstrate that there is no cat in the box after having removed the cat from the box. That's magic!


Ekkehart Schlicht - Disambiguation
July 25, 2012 - 08:23

The co-editor's phrasing you cite -- that the bulk of the related literature bases the analysis of Ricardian expectations on assumptions ruling out that households base their consumption decisions on a disposable income which includes interest income on government debt -- is ambiguous and perhaps misleading. A proper statement would ...[more]

... be: "The bulk of the related literature implicitly bases the analysis of Ricardian expectations on assumptions ruling out that households base their consumption decisions on a disposable income which includes interest income on government debt". My paper made that implicit assumption explicit. This was not common knowledge, as the co-editor's phrasing may suggest.


Ekkehart Schlicht - thank you all
July 21, 2012 - 06:22

I wish to thank the co-editor, the referees and the readers for their comments. All this helped me to understand several problems better and led to a version which, I hope, is more convincing, but obviously was not convincing enough for the co-editor and the referees.


Ekkehart Schlicht - Required Reading
December 27, 2012 - 18:11

For those who are interested in the matter, this seems to me to be required reading:

http://www.britannica.com/EBchecked/topic/165882/disposable-income

http://en.wikipedia.org/wiki/Asch_conformity_experiments


Ekkehart Schlicht - The paper has been published elsewhere
June 06, 2013 - 16:26

In the meanwhile an updated version been published in Metroeconomica 64:3 (2013), (doi: 10.1111/meca.12017, http://onlinelibrary.wiley.com/doi/10.1111/meca.12017/abstract).

The article is not openly accessible, unfortunately. If you are interested, you may request a copy from me (schlicht@lmu.de, subject: Ricardian Equivalence) or consult the most recent pre-print version available at http://epub.ub.uni-muenchen.de/14847/.


Ekkehart Schlicht - I was completely wrong
October 17, 2013 - 06:52

I was completely wrong on this matter. As the first referee noted correctly (and I did not understand), disposable income is endogenous to the problem. I have written an erratum note that I attach.

I can only offer my apologies to the readers, the referees, and the editor for ...[more]

... any unnecessary work and inconvenience caused by my mistake.

Ekkehart Schlicht