The piece of the puzzle you guys seem to be omitting is WHY do yuou 
save?  The old nineteenth century models made it a virtue, an issue 
of morality.  You SAVE because you are sober and thrifty.    
     But that is not why you save.  YOu save either because you 
can't spend it all ... or you save because you can EARN through 
INVESTMENT.  It is the latter that is a good thing for the economy. 
     Well, what you can earn on an investment is VARIABLE.  It 
depends.   
     yes, certainly, creating an environment that "encourages"  
savings is going to perhaps increase investment -- because it indirectly 
reduces the cost of investing, hence raising the return. 
     But -- another way to increase investment is by making it 
more profitable to invest.  Voila!  Instead of going about it 
indirectly -- trying to encourage peo[le to save because through 
the magic of financial intermediation, or through direct translation 
of saved income to direct investment (much more rare), saving does 
become investment -- you make investment itself look attractive 
BECAUSE THERE ARE CONSUMERS OUT THERE WANTING TO BUY PRODUCTS. 
(or services).   
     This way, what drives the investment is what people want MORE 
of.  Gee, isn't that a good idea? 
     Now, there are serious problems with this in terms of conceptualizing 
why it clogs up in the first place, not to mention that this kind of 
simplistic consumption + investment + government = total income model 
is the macro version of the old two-dimensional static micro models 
(like the simple labor-leisure tradeoff model) that works okay as a 
first pass but is pretty far from the real world ... 
     All that notwithstanding ... 
     The lesson from economic history (and solow among others) is 
that what we all THOUGHT "caused" growth may very well NOT be the 
cause of growth.  We all THOUGHT that investment in physical things 
and in "capital goods" "caused" growth.  The first pass at a mathematical 
model of all that in the late 50s was intended to demonstrate it -- 
and it was sort of a shock to all concerned when there was NOT the 
high correlation between long-run growth and savings, long-run growth 
and growth of the physical capital stock, that we had all believed for 
so long.  It was a real "now what?" moment. 
     The significance is in the residuals.  From that point on it's 
all interpretation.  And I think it's a good bet that there's a lot more 
interconnectedness and nonlinearities and multiple possible outcomes than 
anyone has been really willing to model on a macro level (perhaps I 
should say anyone has been ABLE to model on a macro level).   
     But -- I would say that you cannot make a case given the evidence 
and scholarship at this point of time in economics and economics history 
-- you cannot make a case that "savings" should be in any way 
promoted over "consumption".   
     Ironically, by doing so you are making precisely the kind of 
arbitrary deductive centrally-driven econonomic policies that  
many of those who cry out for more relief from capital gains taxes 
find so distasteful.   
     -- Mary Schweitzer