Okay. So a couple of
housekeeping things. I do apologize. McGraw Hills has been
a little bit of a nightmare for a lot of
people that it wasn't working for. If you
clear your cookies or use a different
browser, that's the way to get the textbook
working. Every time I ask them a question,
they're just like, oh, get your students
to contact us at help. So my ability to
help you is kind of limited. But please
let me know if you're still having problems.
I'll pass it on to them. it's really
important that you can access the textbook and
in a week's time that you can access the
homework as well. So please let me know if
you're still having issues with it. And
the second thing is, a student pointed out
a very interesting point to me and I
wanted to ask you what your preferences
are. I'll put out a survey tonight
to get everyone's official opinion,
but I'll bring it up now anyway. Because
we have three mid -semester tests
throughout the semester, I can't believe
they schedule these exams at 8pm.
It's quite cruel actually. I do
apologise about that but because of that
the policy is we cancel three classes
for the semester. So I'm going to offer
you two choices. The first one is just
cancel the classes on the day of the exam
or what I can do is on the day of the exam
I can dedicate that class as like a review
session of the three four topics that will
be on the exam that night which I think
might be useful and then we can cancel three
classes on like a Friday. So yeah, I
see a few nods in the other class as well,
like just before spring break and some other
times as well. So think about what you'd
prefer. I'll ask you. I am a dictator
at the end of the day, so I'll decide
what I think is best, but if there's
a larger majority for one than
the other, I'll definitely take that
into consideration. Any questions about
either of those things? Okay, fantastic. So
we're going to do the first cahoot today the
first official graded one so it's great
that you all rocked up to class and we'll
finish off topic one and get a quick start on
the second topic. I did want to just briefly
go back over marginal analysis because I
think it is so important in terms of how we
think as economists. So when we make
decisions we want to you know evaluate the costs
and benefits of an action but the change
in the action as well. So if you know someone
comes up to you and says hey you know we're
producing 200 units of a good that we're
selling our profit is currently a hundred
thousand dollars are we doing the right thing
the answer to that question is I I don't
know because you don't know if they're producing
the right amount because you don't
know what the marginal benefits and the marginal
costs are and I gave this example regarding
pizza last week that if eight slices of
pizza bring me a hundred units of happiness
on average once again we don't know if I'm
eating the, you know, optimal amounts of
units of pizza. It could be the case once I get
to seven, the eighth one actually brings
me less happiness because even though I
can still somehow stomach it, it increases my
chances of getting sick and I'd be better
off stopping at six or stopping at seven.
So the average profit or the average utility
doesn't tell us anything about how much of
something we should do or how much of something
we should produce. It's the marginal
benefit and the marginal cost of an action
that can tell us that. So we have this
example here, mathematically speaking,
where we have some sort of revenue or
benefit based on the quantity we produce and
some sort of cost. As we could see here with
our cost curve it's Q squared so this is just
an exponential function. As you can see it
starts off with a relatively flat tangent
here but it just explodes over time. This is
what exponential growth this then with our
benefit curve it has this you know interesting
kind of shape where we start off increasing
at some point it is flat and then it starts
decreasing before we get to this point here
a lot of curves in economics kind of look
like this they have this um concave shape and
what this essentially means is it's something
like q to the power of in between 0 and
1 so it could be the square root of q which
is just q to the power of a half and while
this course doesn't use calculus calculus
can help us you know determine a few things
here the first thing is by taking the first
derivative of anything we can see if producing
one more unit results in cost and benefits
increasing or decreasing so for example here
when we take the first derivative of the
benefits of quantity we get the marginal benefit
which here is q equals 250 minus 8q I went
over last week how you get this if you're still
struggling a little bit with this let
me know and I can go over it with you so take
this um if I go here so if we get to an image this is actually
I saw one here which is really good if I can get to
a view image. Okay, this is
tiny, but you have a curve like
this. Actually, I wonder if there's
a way I can draw. Is there like
a function here where I can draw?
Because that would be really
useful. New slide. Oh, that's moving in any
meaningful way. Sorry? Yeah, I could actually. I have a Mac, so is it... you know what can
everyone just kind of see that curve here like
you know what we'll do is imagine imagine here
this curve like stops at the dock so what
this kind of is we could write this as x to
the power of a half or square root of a half
and when we take the derivative of that
what we get if I oh my god what a nightmare
this is I have no idea what's going on right
now so we have x to the the power of 0.5 oh
my god x oh you know what I give up on that
I'm going to do this like not written down
I'll write it down for next class but this
curve here can show us a few things if we stop
here let's say if it's x to the power of a
half it's going to look like that if we take
the first derivative we're going to get a
half x to the power of minus a half this is a
positive number because out the front is a
half so that means that this function as you
can kind to see here every time we increase
the quantity it's increasing the marginal
benefit however if we take the second derivative
so if we derive this a half x to the power
of minus half again it's going to be minus
a quarter x to the power of minus three
over two now the sign out the front is negative
this is the second derivative and what
this tells us is is it increasing at an increasing
rate or decreasing rate because it's
negative it means it's increasing at a decreasing
rate so what this means is the first time
we produce from 0 to 1 we might get a marginal
benefit of 10 when we produce another unit
it's going to increase our total benefit but
it's increasing at a decreasing rate so
it might only give us 9 units of benefit then
8 7 6 5 4 so on and so forth until you
basically get no more benefit from producing
so that's what the first derivative and second
derivative can tell us about a curve's marginal
benefit and marginal cost. The third
thing that we can use calculus for is with
this curve here at this point the tangent is
going to be equivalent to zero and whenever
you get your or you set something to zero when
you take the first derivative you're
looking at either the the local maximum or the
local minimum depending on what your function
looks like. This is essentially the top or
the bottom of the curve and in economics we set
things to zero because that's when we can
optimize for profits so going back here all
the way back here we have our marginal
benefit and marginal cost curves when we make
them equal that's when we maximize our profit
so when marginal benefit equals marginal cost
we can solve this it's when quantity
equals 25 and as I showed last week this is
when the slopes of the marginal benefit and the
marginal cost are the same so if we were
producing here for example you can see that we
start off with a steeper marginal benefit
tangent and then as we move up the curve it's
going to flatten and flatten out. This is
the idea of diminishing marginal benefits or
the second derivative being negative. And
it's the opposite of the marginal cost curve. It
starts quite flat and the slope increases
and increases. So our marginal benefits
decrease as we move and produce more and our
marginal costs increase. So the start the marginal
benefits are much higher than the
marginal cost. They'll be equal here and every
point on the marginal cost will be higher
than the marginal benefits so we can
look at that in a different way so the net
benefits here is just the the marginal benefit
minus the marginal cost and we can
see our quadratic function here and we
maximize profits at this point where the tangent
is equal to zero so this is what i was
talking about before what calculus can
tell us when we have a profit curve or a net
benefits curve it's the exact same thing
we know profits are maximized when the
tangent is equal to zero, or just at the
very top of the curve. Finally, the other way
we can look at it is this intersection of
marginal cost and marginal benefit. So at this
point here, we have our marginal cost curve,
we have our marginal benefit curve here,
and at this point here, you can see the marginal
net benefits are maximized. If we produce
any more, we start losing money, and if
we produce less than this, we're leaving
money on the table even though it's something
like a dollar we're still leaving a dollar on
the table so these three graphs show different
ways of this point of maximization occurs
when marginal benefit equals marginal cost
the slopes are the same we see that profit
is maximized at the highest point and it's
actually the intersection of these two curves
so these are three different ways we can
think about profit maximization in terms of
marginal benefit equal marginal cost and I'll
tell you now a lot of people will get it wrong
by saying And marginal benefits should be
greater than marginal costs to maximise
profits. That's wrong. Just think about the
fact that you're leaving money on the table by
not producing enough. So this is all kind
of just rehashing what I just
talked about. And this is like the
dummy's guide towards producing
goods as a manager. Should you produce
more? Thumbs up if your marginal benefit
is greater than marginal cost. Should
you produce less? Yes, if your marginal
cost is greater than marginal benefit. you
want them to equal each other essentially any
questions about that yeah it can be a whole
bunch of different shapes sometimes marginal
costs could actually be decreasing because
of economies of scale and then you
just produce as much as you can really like if
the benefits are still high some point your
benefits might be zero because you can't
sell any more products costs can't really get
below zero by definition so yeah shapes will
differ that's just an example for you or
when we in a few weeks time we will talk more
about how these costs can vary etc but
great question okay so there's a lot of things
I don't like about the textbook this is
one thing I think they do really well I think
making data-driven decisions is really
important and I think it's not as straightforward
as we might think so I want to take you
through the intuition of using evidence to
think about you know decision making here and
go a little bit beyond the book in terms of
the difference between prediction and
causal effects so for example you might want
to know what the markets are saying about your
product so you know how much to produce
so how do you find out what the demand function
is so there's a lot of data available
to help you quantify your decisions a ton
of research done by economists publishing
these domains you can look for all of that we can
use your own econometric models based on data
simple regressions can tell you a lot
about the world historically people have
hired consultants to help them i have a lot
of hot takes about consultancy i don't think
i'm going to speak about them right now but
i think a lot of consultancy is actually
redundant already um because of how good a
lot of the the ais are i've been playing
around with um you know claude code all weekend
and i've been blown away about how good it
is i don't know if others use them as well
but why pay for a consultant when this can
get you at least 99% of the way, if not more
than 100% of the way. Finally, we have
simple statistical techniques like OLS,
regression analysis, that we can use to
predict the change in one variable based
on another. Pretty easy to use, and
we're at a point where you can literally
just tell these AIs what you want, and
it'll do it for you. So you don't
actually need to know. I think coding's still
important knowledge, but you don't even need
to know it anymore. so what we have here
is the true regression model so imagine we
had every piece of data that existed in
the world if we had that on every person we
could figure out our why why here is some
dependent variable that we're interested in
it could be the the quantity of goods demanded
it could be um the the average score or
the score that someone gets in this class it
could be anything and what we have here is
some intercept parameter a we don't worry so
much about that what we care about here is
this b and this x this x is an independent
variable so if y is let's say quantity
demanded x could be price if it's the the average
score you know that someone gets in this
class x could be attendance and what we'd
want to know is how does attendance predict
the scores people get in this class how does
price of a product predict the quantity of
that product demanded that's what our B can
tell us is some unknown population slope
parameter where a change in X will determine
how much of a change there is in Y and E
is a random error term with these definitions,
don't worry so much about that but
if we had all the data in the world,
our error term would be relatively low if
not zero altogether so the reason why we
have this least squares regression line that
looks like this is in reality we don't have
all the data in the world we don't know
these true effects we're estimating them so
that's why our a and b parameters here have hats
so we might have some data on a market
in terms of the the quantity and price i
might have some data on attendance in in this
class but i don't have attendance on on every
class ever to exist so what we're doing is
estimating an effect for a certain group and
what we want to you know i guess argue about
is can we generalize based on those results
here so we want to know the true b we can
only estimate what it actually is and then
there's like philosophical discussion about it
that applies to the the whole world i guess
so this is kind of what it looks like so
as you can see here this is our our regression
and it incorporates two things one an
intercept and two our b is going to have a
slope so this slope is negative it goes down
like that so b here would be like negative two
or whatever it is what we're doing and how
we get this line don't worry so much about the
econometrics is we're trying to minimize
some of the squared deviations between the
line and the actual data points so you can see
our points occur at different parts and
we want to draw a line that minimizes the
distance of all the points squared combined
that's all we're doing that's what linear
regression is and as a result we'll get this you
know estimate here so for example we could
have you know these 10 observations here and
each observation gives us two things a quantity
of goods demanded and the price of that
good so we have all of these and we can use
regression to figure out how price predicts
change in quantity. So quantity is our
dependent variable y and price is our independent
variable x and we can use that to predict
changes in one and the other. So what you
can do is you can do this in Excel, you can
do it in R, you can do it in Stata, you
could honestly throw this table into GPT
and ask it to run a regression and what it
will spit out is something like this. So as
you can see here, we have our coefficients,
our intercept A, and our price, which is B.
So this is the change in one unit of X. So
X here is price, so a change of $1 in price.
How does it predict a change in quantity?
So the slope here is negative 2.6. so a one
unit change in X will result in a minus 2.6
change in Y. In other words an increase in
one dollar of a good predicts a change of
minus 2.6 units of quantity so people will
demand 2.6 units less. So we can draw it
up like this as well and as you can see these
are our two points. The other thing that
we think about with the regressions is
are these predictions statistically significant?
And without going into a whole diatribe
about p-values and how to think about
them, all a p-value says, really, at the
end of the day, is how likely is something to
occur randomly versus there actually being
a statistical effect from your model. And
we use this threshold of 0.05 usually to
determine statistical significance in
frequentist statistics. So here, the p
-value is less than 0 .05, so we can say
that this prediction is statistically
significant, this correlation between
price and quantity. So, regression
techniques can also be applied in the following
settings. So, we can have different
styles. We can transform a variable by the
log function. When you're working with
wages and stuff, you may want to do that. But
the most important thing here is this
multiple regression. So, think about... What's a good example? Yeah. So, think about class.
as we said before, I want to know what
predicts people's scores on the final
exam. And I could just have one
variable which is attendance. I record
that and predict it. But attendance might not
be the best predictor. There might be other
things that I care about as well. How people
did in the homeworks, how many of the
office hour sessions people came to. I can
measure a whole bunch of different things. And
if I leave anything out, I might not be
predicting the right thing. So I can include
all these different variables so
attendance is x1 office
hours attendance x2, scores on homework
x3 so on and so forth and what this
model will do is show me how much each of
these things predicts someone's overall score
so thinking about what to include and what to
leave out based on data available to you is
an important question okay now we get into the juicy stuff in my opinion so imagine you have
a data set of every person in the entire
world that's ever existed and you know two
things you have all the data on two things
their age when they died and on average how many
glasses of wine they drank a day and we put
this into a regression this is our dependent
variable this is our independent variable
and it shows for every extra glass of wine
a person drinks today on average they live
for 2.4 years longer Based on this
evidence, do you agree or disagree
that drinking wine causes you
to live longer? So, hands up if
you agree with this statement, drinking
wine causes you to live longer
by 2.4 years. Hands up if you disagree with this statement. Okay, majority of
you have your hands up. Sorry, what's
your name there? You? Yeah, Kenneth. Okay, you were
pretty, you know, adamant in that
disagreement. What's your reasoning? Yeah. But the data says
alcohol is good for you. Don't you
agree with the data? No, you don't agree
with the data? Okay, how about,
sorry, is that the hand up there? Yeah,
what would you say? Yeah, great. And let's
go one more. Yeah, you took the words
right out of my mouth. So I should probably
get this tattooed on my forehead so I never
have to say it again. If I had a dollar
for every time I said it, I'd
be a millionaire. Correlation does
not equal causation. All we have here
is a prediction. All we can say is
that someone drinks a glass of wine more
on average a day predicts that they'll
live 2.4 years longer. There's no
causal mechanism between drinking wine
and living longer. An example of this,
well I should say, this is due to
something called emitted variable bias. There's
a ton of different factors that we
haven't controlled for that could determine
the reasons why people drink more wine
and also live longer. So what was your
name again, sorry? Nicole, you know, pointed this
out that people who drink more wine are
probably wealthier it's usually you know
perceived as like a higher class type of drink
and wealthier people also tend to have better
access to health care which could be the
reason that they live longer due to this
better access. So it's because they have more
money that they're able to live longer and
it's also because they have more money that
they're able to drink more wine. So wine has
nothing to do with the causal mechanism here
it could just be a factor due to the you
know people being rich. Another example I came
up with which probably isn't as strong a
reason but wine is drunk drunk probably
more in warmer climates thinking like Napa
Valley in California people in warmer climates
also get more vitamin D I don't think I've
seen the sun here for a few months so
that could be a reason people live longer
vitamin D is important for your health drinking
wine only occurs in warmer climates so
it's not necessarily the wine itself it
could be these other factors now I'm gonna
you know change the question a bit what if
we actually had data on every observable
variable so we had data on people's wealth
their parents wealth their parents parents
wealth etc the weather all these everything
that we can observe we have data on and
we still find that the regression says that
wine predicts people live 2.4 more years
longer on average can we say that now this is
a causal claim that wine causes people to
in 2.4 years longer? Hands up if
you think yes. Still no. Most people
still know. Does anyone have a
confident answer as to what they
think why no? Fairly confident? You could put genetics
into account. Some people's genetics and
the liver can process alcohol better than
others. Yeah, fantastic. So still no. And this
is why you've got to be very, very wary how
you interpret data. So there are also these unobservable variables
that may affect behaviour. So what was your
name, sorry? Christian. Christian as Christian
pointed out genetics is something that is
hard to get observable data on but someone
like me I have one glass of wine and I I
pretty much you know pass out whereas others can
drink a lot of wine for example that's
you know probably a difference in genetics
and maybe due to the fact I'm not 21 anymore but
there are all sorts of things so another
example could be people who like taking more
risks so they have risk preferences to the
extreme are less likely to drink wine because
they're more likely to drink hard liquor they
also take more risk -taking you know
adventures like going base jumping and free climbing
etc so they die a lot younger this would
bear out in the data as people who drink wine
will live longer on average but this is due
to the fact we haven't taken to account
people's risk preferences another example could
be I have you know information on on you
know your final scores in this this subject
and also on whether you attended class it
might not be the fact that I'm you know a
good teacher or anything like that I might be
terrible and you learn nothing here is probably
the fact that people who have more effort
or you know a harder work ethic are more
likely to rock up to class so these are known
as selection effects so these are based
on observable or unobservable traits people
are more likely to opt into something which
makes causal inference saying x causes y
really hard so we've looked at wine but
another example is let's say we want to evaluate
whether this unemployment job program helps
people get a job we can't just measure
like you know the the rights people get a
job in the program and people who don't because
of this selection problem people who you
know more motivated to get a job harder work
ethic are more likely to take the program
they're not so the program might not work at all
it's just that the people in the program
originally had this harder work ethic as
a result how do we measure something like
this and economists care a lot about causal
inference we want to you know implement
policies at work and we want to advise businesses
to do things at work so how do we spend
so how do we go about solving what we call
this fundamental problem of causal inference
what we do to avoid these selection effects
is harness the power of randomization by
doing things like running experiments for
example take this job program what we want
to know is does this program increase the
likelihood that an unemployed person gets a job so
what we can do let's say there are 10 000
unemployed people in the population we
can take each person and we flip a coin if
it's heads we put them in the program if it's
tails we don't put them in the program and
what this means is the way they've been
assigned to the treatment of the job program
or the control of no program has nothing to
do with their own traits, not due to a hard
work ethic, genetics, wealth, anything.
It's due to a random coin. What that means
is we find a way to solve this problem
of selection effects. These two groups on
average should be the exact same in
terms of their traits and their observable
characteristics. So now that we've
assigned each one to the group, now we can see
does the job program work or not because any
difference should be due to the difference
in the treatment manipulation, which in
this case is the job program. So if
anyone's, you know,
read Freakonomics or Nudge or seen
any of these things, this is how they go
about their research. They use this random
assignment mechanism, literally a flipping
of a coin, to evaluate the causal effects of
different treatments. Okay, that's all I
have. I know I went on a little bit of a rant
there, but I think that's really
important. We're going to do our first kahoot.
There's going to be six questions.
Let me get it up. Oh, that was a whole
disaster. So we are 3.30. So remember you
get 0.5 points for an incorrect answer,
so it's always better to answer
than not answer. Of course, what
are we playing for besides extra credit? As per usual,
the winner gets there's a choice between a kangaroo and a koala or a chocolate
bar or a candy bar as you guys say,
of your choice and while we wait,
any questions about marginal
analysis and making better decisions
with evidence? yeah, there's a whole
formula that takes into account the
average and the standard deviation
between your observations Is that the coordination
convenience? Yeah, yeah, there's a whole thing. Yeah, yeah. I mean, the
interpretation's more important, though. Yeah. So p-values with
correlations, I think, are
less meaningful than with causal
work as well. But, yeah, it's good
to know, like, how well something predicts
something. For example, like, sometimes you
don't care about causal inference. You just
want predictions. So in that case, that's
when it matters a lot. Are we waiting on anyone? Are we all logged in? okay let's get let's
get started the overall goal of
the firm is to have fun maximize profits
create your products do the right thing
look sorry okay nice pretty much everyone
got that one right um a good start even
though I think we should all be having
fun the goal of firm to maximize profits okay
i reckon this one's going to be pretty
tight up the top the leaderboard if someone
pulls away a bit i'll probably elicit your
confidence levels of winning which is something i
like to do next one there is no such thing
as a free lunch true or false hey everyone
got that right fantastic i don't even
need to explain it still close up the top this
is going to be a time crunch for sure which
of the following is not one of the five
forces of profitability entry power of sellers
industry rivalry or sunk costs this
one's a little bit more technical in terms
of definition best people got that right
it is sunk costs we did discuss sunk costs
a lot but it's not one of the five forces
okay we've got chicken in the lead by what is
that is that 17 points not much not enough
for me to call you out in class yeah paying
someone more money to do a task will always
incentivize them to exert more effort in
that task true or false fantastic so yeah we
looked at quite a few examples from papers
we're actually giving people money to do
something decreases effort in certain
circumstances we have to be very careful how we
put out incentives oh yeah we have a new
leader who's I'm cool frog cool frog what's
your name welcome to the leaderboard Michael
well I should say welcome to first we've
got two questions left. Any confidence
in yourself to hang on? Nah, fair enough. It is so tied up here. Yeah. The equation for present value is as follows. PV equals FV
divided by 1 plus R of the power of
N. True or false? Fantastic. It's
true. I'm not trying to trip you up.
These are fairly straightforward. If you
pay attention class, you should know
what's going on. Okay. One question left.
You're up by 50 points. I don't know
how that converts to seconds in
Cahoot actually. Do you back
yourself in for this last one now?
Okay, amazing. Okay, best of luck.
The person who was in polling
position last class got the last
question wrong. So hopefully this
doesn't happen again. A firm maximizes
their profits when marginal benefit
is greater than marginal cost,
marginal benefit is smaller than marginal
cost they equal each other or none
of the above okay great so in the in the
previous class a third of the people said
marginal benefit greater than marginal cost
remember you're leaving money on the table
with marginal benefits greater than marginal
cost don't pack up your computers I still
have 15 minutes of your time but let's see
yeah you back yourself congratulations you
can choose between little kangaroo and
the koala package deal or a Snickers
or a Milky Way? One of a kind. Nice. Congratulations
and there's going to be 11 more
opportunities. Oh there's a little
koala in third there. Hilarious.
Okay great. So let's get on to the the second topic here. So we're going to talk
about market forces. So this is going to be
kind of a summary of the laws of supply
and demand and the implications but we're
also going to extend the model a little bit
further so if you haven't taken econ since high
school you forgot everything about your
courses you took last year last semester
don't worry I'll refresh you and for those of
you like done this before we'll extend it
so we'll talk about the law of demand the
law of supply and the implications of that
we're going to speak about how we calculate
surplus and surplus is how we determine whether
something is efficient or not in economics
so we can talk about whether that's a good
measure or not but I'll show you how to
do that we're going to look at how price and
quantity determination in a competitive
setting and comparative statics i might play a
game with the class as well to show you how
this actually comes about in a certain
environments and we're going to talk about
taxes both the excise and the ad valorum and
how price controls can you know distort markets
so i i said before there are a lot of
things i don't like about the textbook one of
them is i am you know a big advocate for
free markets. I think they're great at allocating
resources. However, they don't work all
the time. There are situations that we know
of that markets can fail and they decided
to put this chapter all the way in the appendix.
So what I'm going to do is take that out
of the appendix and put it at the end of
this chapter, maybe have it as its own chapter
where we talk about things like externalities,
public goods, information asymmetry
and learn when markets are the best way to
allocate resources and what we can do to
potentially solve that. So our definition of
demand is just a maximum quantity consumers
are willing and able to purchase at various
prices and we'll end this class on an actual
example of that. So we can represent demand
as either a table of data which we actually
saw in the previous lecture slides, a graph
which we'll look at in a second or an
equation and we saw the regression of a demand
curve earlier. So previously you've probably
seen it as like the quantity of demanded
equals some function of the price of X,
whereas the prices of X changes it's going to
affect the quantity demanded. So for example
we could have price equals a hundred minus
two times quantity or if we care about looking
at it as quantities our dependent variable
we can have the quantity of X equals fifty
minus a half of price. And this is an inverse
relation so this means as price increases
the quantity demanded is going to decrease
as this is essentially the law of demand the
quantity of a good consumers are willing
and able to purchase increases as the price
falls which means they're inversely related
so the market demand curve illustrates this
relationship between total quantity at a
specific price per unit of a good all consumers
are willing and able to purchase holding
other variables constant so it looks like this
so you can see here when the price is
really expensive at $40 for this good no one
wants to purchase it when the price is zero
there's 80,000 units that are demanded it's
free a lot of people want it and when it's
somewhere in between you have some people
that are willing to pay for it others not so
at $20 you have 40,000 units of quantity
demanded so this price falls more people are
going to get benefits from purchasing it because
everyone has a cost that they're willing
to pay as soon as it falls below that you're
willing to pay it. So we're going to
end up on this little game. So I'm going to
offer you all a service. How much would you
be willing to pay me for me before class
on Friday to go to Leaps, the coffee shop,
purchase you a coffee or any drink of your
choice and bring it to you and serve it
to you here in class? So how this is
going to work is I'm going to
start by offering this service for
a price of $0. Everyone who wants
this service for $0 is going to stand up
i'm going to start increasing the price
and if you still want to do it at that price
you stay standing if you're like no i
don't want to pay that sit down and i will
buy you this and bring it to you but you're
going to pay me the money as well yeah
including the coffee as well everything
included all all yeah everything's included
in it so let me just get something i need
to download one thing and then we can do
this so if everyone who wants this service
at zero dollars, you pay zero dollars,
please stand up now. And I assume everyone
will stand up. If you don't, I'm also
happy to buy you like a little baked good
or something as well. Whatever you want.
One thing from Leaps. Okay, so everyone
standing from the Cahoot, I think we have 50
people in class, so 50 people are willing to
do it at zero dollars. Okay, would you be
willing to pay me 10 cents for this service on
Friday? Does anyone not want it for 10
cents right let's keep everyone up 25 cents
is anyone out still everyone great 50 cents
still fantastic one dollar everyone still
standing two dollars fifty no one's out for two
dollars fifty okay we've got one out
we've got one out okay we've got 49 now four
dollars four dollars okay we've knocked out
quite a few people so I reckon 20 people have
gone so we're now 29 let's go 36 dollars
okay we lost what 10 more so let's go 20
$8 50 okay how many we have the weekend count
now are you still standing at the back
there one two three four five six seven eight
we've got eight people now eleven dollars is
anyone still standing for $11 up to you you
can sit if you want or you can win right
now on your own for $11 okay congratulations
you won for $11 what's your name Michael come
up to me after give me your order and I'll
bring it to you on class on Friday what
we've done here is not just play a game where
I earn a little bit of money but also what
we've actually done is we've sketched out
our own demand curve I know it's a funny
-looking shape but this is what a demand curve
essentially looks like as the price is
really low everyone wants it everyone's demanding
the good but as the price increases at
some point you're like yeah it's no longer
worth for me to pay that price so less and
less people are going to want the good as
the price rises all the way up until we got
to $15 where we're at zero so this is what
a demand curve is this is how we can derive
it we want to see what is the most a person
is willing to pay for a good or service And
that's what you just all calculated in
your own heads. And through this aggregation
of these willingness to pays, we're able to
determine the quantity demanded at a specific
price for a good. That's all the
demand curve is. Okay, is there, sorry,
is there someone in my class who wanted
to do a call-out or something like that? Or
is it in my next class? Must be my next
class then. Alright, great. No, that's all I have today. One thing I will say
is there actually is a call-out for
the QBE group, which I'm the academic
advisor of tonight. I'll get this out
here, call out. Why is it in Arabic? What did I do? I wonder if it's
still not right. Okay, well, it's
in Kranit somewhere at 6 o'clock. I spent
$150 on insomnia cookies, so they'll
have that there. I clearly have done
something to this computer, so I apologize.
I can't give you more information on
that right now. But, yeah, I'll see you all
on Friday. And come up to me and give me
your order. Yeah. I have a question. Yeah,
what's up? If you can go back to the
Excel sheet. Yeah. So in this case,
wouldn't your quantity be the independent
variable and price be on the X axis? So
price is on the Y axis. Well, because in this
case, you're quantity dependent on the
price, right? I mean, you can technically
do it either way, but the way we usually
do it is price on the Y axis, quantity on
the X axis. So if you look at the slides
here, we go back to here, you see we have
price. Yeah, that's just how we would model
it. Technically it could have ended up
being the other way around. I was just
confused because in this case your quantity
of people standing up dependent on the price.
Yeah, yeah, yeah. So like usually we
know what a price of a good is. That's why
when we run regressions, quantity is the
dependent variable. But when we
graph it, we do it this way.
Yeah, no worries. Hi. Hey, how's it
going? It's going great. So the slides, I was
wondering if they're like fully updated
in the Bridespace. Okay, so the from the
first section notes redacted and I wrote
these new ones last night So I'm put
them up yet. In fact, I actually haven't
even finished writing up the lecture So
I'll do that and put it all up for you.
Okay. Yeah, no worries No, no, no second
50th. You're all losers to me. So yeah,
but you'll have a little more opportunities
Yeah, but it'd be got to be quicker
on the trigger But remember it's about
the extra credit at the end of the day anyway,
so second is pretty good in that regard
Hey, how's it going? Have you ever honours contracted a
class before? I'm doing it for the
first time for my experimental class,
actually. Okay, would you be open to doing it
with me for this class? Yes, I'm open to
it, but I genuinely don't know what I've
assigned to you. Do you want to email
me, and then we can meet, we can sit down,
and we can discuss options. Perfect,
thank you so much. Have a good
one. No worries. Hey. Hello. What was your name
again, sorry? Michael. Michael, okay,
congratulations. also if you're willing to pay nine
dollars like I'm happy to do that for you
okay what do you want a 20 ounce flat white
Michael 20 ounce flat white and can I also
throw in if there's money left over can I
throw in a pastry as well sure dude do you
have any preferences whatever I got so and in
terms of payment I like you can transfer to
me like light advice they'll also only after
I give it to you as well I'm not taking
any money okay but yeah congratulations yeah
no and thank you for illustrating the point
as well that means a lot to me yeah
yeah I screwed up in that regard I didn't
think about that yeah maybe I should pay
you all right cool I'll see you on Friday
Michael yes how's it going I'm
exhausted yeah yeah look to be honest
though I taught 730 I am last semester
oh yeah oh my god my condolences yeah Do you
follow Premier League? I do. Coach your team.
I'm a huge Arsenal fan. I was just there
a few weeks ago. I went up to watch some
games. Oh, awesome. How many games
did you go to? Two. We saw when they
played both ties. So did you go to
like the Tottenham Stadium? Yeah, it
was really nice. Yeah, I went there
actually. So I lived in England for
the past few years before moving here.
But she went to one of the NFL games
at Thompson State. Yeah, no, they're good. Yeah. But, yeah, every
time I go up to London, I
get to a couple of Arsenal games
as well. So, yeah, no, it's
really good fun. How good is
the atmosphere? It's awesome. Yeah. Do you watch football
here as well? Yeah, I'm from
Houston, so I'm a every Houston team. Texans.
Oh, nice. Nice. We caught our L. Yes, that was, I've
never seen Stroud play so bad in my life.
Oh, he's, I don't, he's seen Go's.
I don't know, he just needs his
confidence back. Yeah, yeah, it was weird. Because, like, May
fumbled it, like, five times as well.
Nine turnovers in the game.
Unbelievable. And there should have been
more as well, if, you know, you
could form a ball. We've got a sad
Bears fan back there as well. I'm so
sad, man. That was the worst game ever.
That might have been the greatest
throw I've ever seen in my entire
life. Yeah, yeah. I swore when that
pass, when you caught that, that
the game was over. So, I was downtown
at, what's it called, the Bears
bar at Kirby's. it's the guy who owns it's a Bears
and Cubs fan. I don't know if
you've ever been down there in
Lafayette, but I'll see him with a
couple of Rams fans and everyone else
at Bears fans. It just went
absolutely nuts when he made that throw. I had a bunch of
people over at my place and
they all just do things like
that. Nice, yeah. But you must be
excited for the future, though, with Ben
Johnson. Yeah, I think we've got a
good five years. Yeah. For sure. Yeah, my mate's a
Packers fan. I've never seen him so upset
in his life, yeah. Unbelievable. And he made an
incredible fourth down throw that game as
well. That's insane. I still find it
funny how we beat you this year with Tyler
Huntley, given all the miraculous wins that
you have as well.