Revision for EASE paper.
1. Comments from editors and responses pasted below:
2. This list is of changes we plan to make, aside from the suggestions:
a. Make sure we have made no policy recommendations at end. This is just a
matter of removing “should” and such words, not of changing the
substance of our discussion. (Policy discussion in conclusion revised to
eliminate recommendations.)
b. Add arrows to Figure 2 as in ppt, for Asean, Japan, India, and China.
c. Add a triangle plot for how old age is provided for in different countries,
together with a discussion of it. (Added as Figure 9, discussion on p14;
renumbered old figures 9-11 and text references)
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Date: Sat, 21 Jun 2008 00:42:21 -0700
From: Andrew Rose <arose@haas.berkeley.edu>
To: "rlee@demog.berkeley.edu" <rlee@demog.berkeley.edu>
Cc: Andrew Rose <arose@haas.berkeley.edu>,
"ITOINTOKYO@aol.com" <ITOINTOKYO@aol.com>
Subject: EASE-19 Proceeding with Publication
Parts/Attachments:
1 OK ~93 lines Text
2 Shown ~472 lines Text
----------------------------------------
We are writing to you with details concerning the publication of the conference
proceedings from EASE 19 that was just held in Seoul. We plan on submitting a broad
selection of the papers to UC Press for publication in a conference volume.
We would
like to have your revision by Friday August 15; please let us know immediately if you
will have any serious deadline hitting that target. We would rather drop a paper or two
than have our publication deadline lag seriously. Of course, we make no advance
commitment to publish any given paper, and reserve the right to request further revisions
in any case.
By August 15, we would like to have your paper, formatted in standard Univ of Chicago
Press format (details are available at the NBER's website) in Word, via e-mail. You
should send it to: a) both of us (arose@haas.berkeley.edu and itointokyo@aol.com); b)
your discussants; and c) Helena Fitz-Patrick at the NBER (hfitzpat@nber.org). Please
also enclose a brief statement detailing how your paper has changed from the version
distributed at the conference, and in particular, how you have changed the paper to
respond to the comments made at the conference and by us. The revision will be posted
on the conference/publication website, and you're free to submit it to the NBER Working
Paper series. Please ensure that your paper includes an abstract; if you didn't submit one
originally, please send your abstract now. Also, please ensure that there are no policy
recommendations in your paper, consistent with NBER guidelines.
In order to speed up the process, we have a number of suggestions for how you might
want to consider revising your paper before submission. We think that the most
important things you should consider in your revision (above and beyond the comments
raised by the discussants) are as follows:
Are bequests relevant for economic life-cycle? Especially for income of elderly, and
especially for issue of transfers vs. savings. [Added statement on p 14/15: Bequests are
potentially an important motive for saving, in addition to life cycle saving, but neither we
nor anyone else knows how the motivation to bequeath varies with fertility. Also
discussed in the context of the discussion of the triangle graph.]
Transfers vs. Savings: the savings aren't very high (one generation's consumption).
[Added to p 15: That US elderly depend more on assets for their retirement than
Japanese elderly may seem surprising given that Japanese saving rates are higher than
US saving rates. The estimates are constructed to insure consistency with NIPA
estimates of saving. There are many possible explanations of the seeming
inconsistency, however. This is a snapshot of the use of assets to support retirement by
the elderly. Aggregate saving rates also depend on the saving behavior of non-elderly
adults. Moreover, the estimates presented in Figure 9 address only the lifecycle use of
saving and not other motives, e.g., the bequest motive.]
An issue that current economic life-cycle may be irrelevant in future with longer life and
therefore longer working life? [Added on approximately p10. “ An important issue is
whether the economic lifecycle is changing over time, and how, and whether it is
susceptible to policy. The potential for policy intervention is discussed in the conclusions
to the paper. An important possibility that is widely discussed is that the age at
retirement will increase as health improves and lifecycle expectancy rises. This may
happen and in a few industrial countries including the US, labor force participation rates
have risen slightly at older ages in recent years. For the most part, however, the age at
retirement has declined dramatically around the world. For example, the median age at
retirement of US men was above 74 in 1900 and is now around 63. This is a typical kind
of change for industrial nations, although Japan has been different. The labor income
profile for the US and Taiwan have become increasingly concentrated declining for both
the young and the old relative to prime age adults. Another possibility is that the
consumption side of the economic lifecycle will change. Our preliminary analysis
suggests that this may be occurring – consumption is rising most steeply at older ages in
the industrial countries and particularly in the US consumption of the elderly has
increased very sharply. In the analysis presented below we abstract from these changes.
If we were to incorporate them, however, they would reinforce our conclusions.”]
Should one worry that support ratios are basically always below 1? [No. It is to be
expected that they will be below unity since it is typical that aggregate consumption
exceeds aggregate labor income. All that really matters here is the proportionate change
in the support ratio. Revised text on p 11 reads: The impact of the economic support
ratio does not depend on its level; its effect on income per effective consumer – the first
dividend - is determined by the growth rate of the support ratio (Figure 7).]
Is it worth pursuing causality via IV for the fertility-human capital relationship? [That
would be a different project. We hope to do that in the future, but not for this paper. No
changes made in text, but we could add a statement that it would be important to pursue.
We do make clear that the causality is an important issue that we have not addressed.]
(Finlay) Youth-age and Old-age dependencies are very different: a good point! [Agreed;
addressed in the introductory paragraphs. For example, “Although children and the
elderly are both referred to as dependents, they differ in a very important way. Children
rely almost exclusively on transfers to the fill the large gap between what they consume
and what they earn. The elderly, in contrast, rely on a combination of transfers and
lifecycle saving to fill the gap between what they consume and what they earn. Thus,
aging – and the anticipation of aging – can lead to an enormous increase in transfers
and/or assets.”]
(Finlay) First dividend is mechanical. Increase working age share WA/P, then increase
Y/P. The working age share is higher, if the income/worker does not change, then percapita income increases. [Agreed, made revision to paper on p 10 or so that reads:
“Note that equation (2) is an identity and, hence, given output per effective consumer,
changes in the support ratio must produce point-for-point changes in output per effective
consumer. A comprehensive understanding requires that we explore the second
channel, as well, how changes in population age structure, other population changes,
and non-demographic factors influence productivity growth, i.e., the growth of Y/L.”]
(Finlay) Second demographic dividend from fact that capital accumulation increases with
longevity (transfers don't deliver it). Although it is acknowledged in the paper, second
demographic dividend is not straight-forward. Capital accumulation increases K/L. [No
changes made - A major chunk of the paper is about the second demographic dividend, a
term that we ourselves coined in earlier papers. It depends on much more than increased
longevity, as we currently discuss. And it will increase K/L only to the extent that the
economy is closed. But all this is already discussed, we believe.]
(Lee) Are the dividends from the second transition really that big? [Inserted: “A note of
caution is in order here. The very large simulated effects are conditioned on low
intergenerational transfers. The evidence from Japan, Taiwan, and preliminary
estimates for Korea not presented here indicates that intergenerational transfers to the
elderly were closer to the high IG transfer scenario than the low IG transfer scenario.
Hence, the second dividends realized in East Asia were probably well below the possible
gains that could have been realized. Estimates of the contribution of changes in age
structure to growth in per capita income in East Asian range from about one-third to onequarter of actual growth (Bloom 1998; Mason 2001). The simulated impact on
consumption is substantially greater than the simulated effect on per capita income,
because of the rapid increase in consumption rates for Korea, Japan, and greater
China.”]
(Ito) Baby booms after war: can they fit in? [Inserted: “The changes in age structure
that accompany the demographic transition are emphasized here because of their
importance particularly in countries which have experienced rapid fertility decline in East
Asia and elsewhere. The post-World War II baby booms of the US and other Western
countries have also produced significant changes in age structure and demographic
dividends. These are qualitatively similar to those experienced in East Asia but smaller
in scale. Although Japan experienced its own post-war baby boom, it was short-lived
with little discernable effect.”]
The mysterious triangle (one of the slides) may/should be added to the paper. It is
surprising to see that Public transfers large in US, larger in Japan and Family support,
Japan zero. [Triangle graph added with discussion of its implications on p 14. The
discussion does not attempt to explain why the US and Japan (or the other countries)
differ. An interesting topic we expect to explore in the future, but not particularly
relevant to this paper.]
You may want to touch on the issue of childless family, and whether saving and asset
accumulation and dissaving-you mentioned Mike Hurd's work? [Inserted the following
“Hurd (1987) finds that the bequest behavior of individuals with and without children is
similar, suggesting either that bequests are mainly unintended or that saving for bequests
may be motivated by a others besides ones own children (Michael Kuehlwein (1993).”]
Fig 9 - 11. In the long run, Low IG is better in the long ran than high IG. [This point is
addressed on p 17. Inserted: As compared with 1950, changes in age structure lead to
about a 30% increase in consumption per equivalent consumer in 2030 given the high
IG transfer policy (Figure 12). Using the low IG transfer policy, changes in age structure
lead to an increase in consumption per equivalent consumer of about 50% in 2050.
Note that the higher consumption after 2025 for the low IG transfer policy comes with a
cost. The higher saving rates and lower consumption rates necessary lead to lower
consumption between 1995 and 2020 under the low IG transfer policy than under the
high IG transfer policy. Consumption remains permanently higher under the low IG
transfer policy. Over the next 100 years (not shown) consumption is 20 percent higher
on average given the low IG transfer policy. In a closed economy these differences
would be larger. ]
How are the different age groups affected by the high IG and low IG regimes? 65+ still
saving, still hold assets? 65+ increasing consumption e.g., Japan. [Added on p 17: The
simulations presume that the economic lifecycle itself does not respond to changes in
age structure and, hence, the gains (or losses) in consumption are equally shared by all
age groups. Of course, other outcomes are possible. The elderly might flex its political
power and increase its consumption relative to younger generations. Or young
generations may rebel if IG transfers are too burdensome to the detriment of the
elderly.]
(This is the union of in-conference notes taken by us; we apologize for any redundancy or
lack of clarity in them.)
Around late Autumn, we anticipate sending you two referee reports (one from the NBER
and another from UC press). At that point you will be given another short while to revise
the paper, taking into account the views of the referees.
We thank you again for your participation in EASE 19 and look forward to receiving the
revision of your paper.