[ExI] Psychology of markets explanations

[Dan]

--- On Mon, 6/1/09, BillK <pharos at gmail.com> wrote:
> On 6/1/09, dan_ust wrote:
>> No so.  This is not a perfect law in the sense
>> that it doesn't apply.  Instead, it
>> applies to all such phenomena.  To wit, armed
>> with this law (and other economic
>> laws) one can look at the data and attempt to
>> interpret it correctly.  This is
>> especially important in cases where it appears that
>> the law is being violated:
>> it directs you to look for confounding factors. 
>> (Of course, at the limit, it should
>> call you to question the law; economic science has
>> progressed by steadily
>> weeding out "bad" economic laws -- I mean laws are
>> either are completely false
>> or only apply to a limited range of economic
>> phenomena.)
> 
> I think we are just disagreeing slightly about what a 'law'
> is exactly.
> 
> I say that if it is a law that cannot be used to make
> predictions,
> then it is of little practical use.  i.e. it will lead
> you to make bad investments.

How will it lead you to make bad investments?
 
> You seem to say, (one can never be sure), ;) that,
> sure, often things
> happen which don't follow this 'law' but that is because
> other things
> have changed. I agree it is nice to analyze why my
> investments have
> failed, but I would prefer to have a 'law' that told me not
> to make these investments in the first place.

Not at all.  My point is that this particular law -- and laws of this type: economic or praxeological laws -- are not the same as other types of scientific laws.  They are more akin to laws in the deductive sciences in that the phenomenon that fall under simply cannot violate them.

In this case, it's not that if you see a price rise and supply has increased, the law has been violated, so we have to make up stuff to get it to work.  Rather, if supply increases and the price rise, then we must look at the demand: the law focuses us on explaining the price rise when this happens along with an increase in supply as a rise in demand.  Then, when we want to explain why the price rose, we have to frame any hypotheses about this in terms of why demand rose.  This is how laws work in economics.  And they don't, as I've said before, have predictive value in the same way, say, chemical laws do -- because there are no constant relations in human action -- at least, none we know of.  What's meant by this last point is one can't apply something like Mill's methods to empirically validate these laws -- e.g., we can't say a 10% increase in supply will lead to a 5% fall in price.

Regards,

Dan