00:00:02
we're here at Wen marketing Professor
00:00:03
Ron Burman and Pablo Hernandez who is a
00:00:07
professor of Economics from NYU Abu
00:00:09
Dhabi they're here to talk about their
00:00:11
paper which is all about the burn rate
00:00:13
of startups welcome thank you Deborah
00:00:17
well first of all can you set the scene
00:00:18
for us what is your paper all about so
00:00:22
um what our paper looks at is how the
00:00:25
burn rate of startups influences the
00:00:28
chances of a startup to go Bank rupt and
00:00:30
burn rate which is more of an industry
00:00:32
term is um how much money does a startup
00:00:35
spend every month per employee so for
00:00:38
example is if a company spends $300,000
00:00:42
every month and they have five employees
00:00:44
the burn rate would be $60,000 per
00:00:46
employee and it's a number typically the
00:00:48
industry uses to compare different
00:00:49
companies and see which ones are are
00:00:51
burning too much or too little
00:00:54
money yeah I wanted to add to that be
00:00:57
that it's a very simple method
00:01:00
because you just have an amount of money
00:01:01
divided by the number of people there
00:01:03
you don't go into the details of where
00:01:04
do you spend the
00:01:06
money and how why is that important to a
00:01:09
startup so there are two reasons why
00:01:11
burn rates are interesting uh and
00:01:13
important for a startup to look at the
00:01:15
first thing is that startups typically
00:01:16
operate on investors money and they want
00:01:18
to make sure they don't run out of money
00:01:20
until the next time they can fund raise
00:01:22
so they want to make sure they don't
00:01:23
burn too much money too quickly and the
00:01:26
other reason is that investors from
00:01:27
their side are trying to figure out
00:01:29
which startups are operating efficiently
00:01:31
but also more successfully and because
00:01:34
the startups are very different types of
00:01:36
businesses the only Apples to Apples
00:01:39
comparison is typically how much money
00:01:41
they can make or they can spend and the
00:01:42
burn rate as Pablo said is a great
00:01:44
metric for that so what are your paper's
00:01:46
key takeaways so the key takeaways there
00:01:50
are there are quite a few but I think
00:01:54
that the most important one is that we
00:01:57
allow or we give to be more precise a um
00:02:03
a notion that if you spend too much or
00:02:06
too little the companies are more likely
00:02:09
to to fail there is a balance burn rate
00:02:13
um on in addition we also look at the
00:02:17
how the entrepreneurial characteristics
00:02:19
in particular human capital are related
00:02:22
to this sound decision making this
00:02:24
balanced uh um burn rate and we find
00:02:29
actually that that one dimension of
00:02:31
human capital which is um education is
00:02:36
positively related to to a balanced burn
00:02:39
rate and and hence uh lower chances of
00:02:43
failure so so just to give an example um
00:02:46
we find that more education actually
00:02:49
reduces uh the chances that a startup
00:02:52
will fail and another interesting
00:02:55
takeaway that we find is that um
00:02:58
something we call confidence or optim
00:02:59
optimism of an entrepreneur actually
00:03:01
helps to start up a little bit um and
00:03:03
this is a unique part of our research
00:03:06
what conclusions of any surprised you so
00:03:09
actually let me first start with what is
00:03:11
not surprising in our findings um a very
00:03:14
well-known result is that if companies
00:03:16
burn too much money they're probably
00:03:17
going to fail and that makes sense if if
00:03:19
you're overspending and if you're
00:03:21
spending lavishly on things the company
00:03:23
doesn't need they might go bankrupt the
00:03:26
surprising result was that actually
00:03:27
under spending money spending to little
00:03:30
also increases the chances of failure of
00:03:32
a company uh we have a few conjectures
00:03:34
why that happens but this is one
00:03:35
surprising result um a second surprising
00:03:39
result is that a lot of past researchers
00:03:42
looked at what's called human capital as
00:03:44
Pablo mentioned and typically human
00:03:45
capital used to be um two things
00:03:48
experience and education we find almost
00:03:52
no effect for experience um adding a lot
00:03:55
of experience to entrepreneurs doesn't
00:03:57
make the companies fail less but we find
00:03:59
quite a big effect uh for Education uh
00:04:02
so there is a big gap between failure of
00:04:05
let's say someone with a bachelor's
00:04:07
degree versus someone with a high school
00:04:10
diploma or someone with a master's
00:04:11
degree versus someone with a bachelor's
00:04:13
degree actually that effect is
00:04:15
significant if we look at entrepreneurs
00:04:17
with a bachelor's degree versus an
00:04:20
entrepreneur with a high school diploma
00:04:23
the there is a 5% more chances than that
00:04:28
the bachelor's degree the entrepreneur
00:04:30
with the bachelor's degree uh succeeds
00:04:33
or doesn't fail than the other
00:04:35
entrepreneur and although so 5% might
00:04:37
seems small this is um like
00:04:39
entrepreneurship right so the data is
00:04:41
very noisy Etc it is a big effect it is
00:04:44
a big effect just from the difference
00:04:47
between entrepreneurs with a high school
00:04:48
diploma to entrepreneurs with a
00:04:50
bachelor's degree so what are some
00:04:52
practical implications of your findings
00:04:55
how can people use this information to
00:04:57
help them so the practical implic
00:05:00
again that you can look at them from the
00:05:02
perspective of the investors and the
00:05:04
perspective of the entrepreneur from the
00:05:06
Investor's perspective is that given
00:05:09
that investors are exposed to uh
00:05:12
sometimes a large uh range of companies
00:05:15
they can estimate how how an individual
00:05:20
a given company is spending with respect
00:05:23
to the others and they can assess
00:05:25
whether the expenditures are balanced or
00:05:28
not uh for R preneurs uh it may work in
00:05:32
a similar way for entrepreneurs they can
00:05:34
go and look at the benchmarks uh the
00:05:36
industry benchmarks and see whether they
00:05:38
are too different too much or too little
00:05:42
uh they are spending too much or too
00:05:44
little compared to what the industry is
00:05:46
doing on average and then given these
00:05:48
numbers for example investors can say I
00:05:51
need to look at a portfolio of 50
00:05:53
companies I cannot give attention to all
00:05:55
of them but I can um kind of signal out
00:05:58
the ones which are very extrem too
00:06:00
little or too much and maybe look into
00:06:01
them and see whether they have problems
00:06:04
and entrepreneurs on the other hand
00:06:06
entrepreneurs are
00:06:07
typically unsure of how much to spend I
00:06:10
don't know if I should spend more or
00:06:11
less I don't know what the competition
00:06:13
is doing Etc if you take those numbers
00:06:15
this metric is very simple to look at
00:06:16
and you can say uh how close am I uh to
00:06:19
what the market is doing and if I'm
00:06:21
Different it doesn't mean it's bad but I
00:06:23
need to have a good explanation for why
00:06:24
it's bad I need to give a good reasoning
00:06:26
so this is the Practical implication so
00:06:29
what said your research apart from other
00:06:31
work in this area so there are a few
00:06:34
aspects uh which make our research
00:06:36
unique uh the first one is the data
00:06:38
source that we're using um this is
00:06:40
confidential data that we received from
00:06:42
the cun Foundation uh which collected uh
00:06:45
for seven years between 2004 and 2011 um
00:06:49
a long survey about a representative
00:06:52
sample of companies that was founded in
00:06:53
the US it's over 3,000
00:06:56
companies um and they asked them to give
00:06:59
information about employees and about
00:07:01
other things um and we're one of the
00:07:05
only researchers that have used this
00:07:07
data in this approach yeah another
00:07:10
another um important departure from
00:07:13
previous literature is that even though
00:07:15
the burn rate is is is a term that that
00:07:17
is used widely in the industry uh there
00:07:20
is surprisingly little little research
00:07:22
at least to our knowledge on that
00:07:24
important metric and we go to the to the
00:07:27
exact exact core of that and and the
00:07:30
final thing is that a research looks at
00:07:32
the Dynamics of the company a lot of
00:07:34
research just says this is the companies
00:07:37
that started and we see how many failed
00:07:39
for us we're trying to see what is the
00:07:41
chances of failure next year and another
00:07:43
year and another year which allows us to
00:07:47
um filter out a lot of causes for
00:07:49
example in our data we have 2008 the
00:07:51
economic crisis of course more companies
00:07:54
were fed during the economic crisis but
00:07:56
we're able to filter that out and say
00:07:58
what is the effect of the B rate above
00:08:00
that um and this gives us uh statist in
00:08:03
statistical language it gives us more
00:08:05
power but we can say better what is
00:08:07
going on with those companies so how
00:08:10
will you follow up this
00:08:12
research we have two major points um the
00:08:16
first one is um try to see whether our
00:08:20
results are robust so we are currently
00:08:23
uh trying to trying to find the same
00:08:27
effects or looking for the same effects
00:08:28
on a data set that's provided by the
00:08:30
government and that is research in
00:08:32
progress and the other thing is what we
00:08:35
were talking about or we mentioned
00:08:36
earlier the the mechanism so what's
00:08:39
what's driving this this balanc or
00:08:42
unbalanced uh burn rate and and I think
00:08:45
well if you can you want so so and we
00:08:48
have a theory so again as I said the
00:08:51
interesting or the surprising result is
00:08:53
that spending too little is bad and
00:08:55
spending too much is too bad like both
00:08:56
sides are too bad and you want to stay
00:08:59
kind of in this optimal Center there and
00:09:02
the question is why do you get those
00:09:03
extremes um and our current conjecture
00:09:06
is that um a lot of it comes from the
00:09:10
fact entrepreneurs are unable to predict
00:09:12
um demand in the future and competition
00:09:14
in the future as a result they need to
00:09:17
plan today for what will happen um and
00:09:20
sometimes they over plan like the
00:09:21
overbuild capacity which just cost them
00:09:23
a lot of money so this is spending too
00:09:25
much and sometimes they're under plan
00:09:27
and when they uh built too little and
00:09:30
there's a lot of demand in the market
00:09:32
they're unable to supply and make the
00:09:33
money they could have um and as a result
00:09:36
other companies are raising the funding
00:09:38
and for this entrepreneur he has other
00:09:40
opportunities and they just decide to
00:09:42
close the business um so this is our
00:09:44
current uh kind of working Theory um and
00:09:48
our next steps are basically to validate
00:09:49
that both using the data and using
00:09:52
theory model that we're try to explain
00:09:53
that actually Ron mentioned a very
00:09:55
important point that is fundamental
00:09:57
which is the opportunity cost
00:10:00
that I we think has a lot to do with
00:10:03
this uh surprising result that spending
00:10:06
too little may lead to failure because
00:10:08
the entrepreneur might say okay I'm
00:10:10
making money I'm not losing money but
00:10:12
I'm spending so much time I have a
00:10:15
probably a lot of Education a lot of
00:10:16
connections I might use that in another
00:10:18
business and that is something that we
00:10:21
conjecture is at the core of our of our
00:10:23
result great well that's it for now um
00:10:26
thank you very much for joining us thank
00:10:28
Youk you very much for having us
