Note: Since these log files are derived from the Closed Captions created during the Channel 6 live cablecasts, there are occasional spelling and grammatical errors. These Closed Caption logs are not official records of Council Meetings and cannot be relied on for official purposes. For official records, please contact the City Clerk at 974-2210.

>> the mayor is out, council

member martinez is at a doctor's appointment with his wife.

If the chair would entertain a motion for us to take a recess 45 and start the 45 whether he is here or not.

That agreeable?

>> Moved.

>> Second.

>> Aye.

We will take a recess for half an hour from now and then start promptly at 10:45.



>> Ok.

Thank you.

Today is work session number 6.

We are today talking about our cost allocation methods.

I'll be doing the first part of the presentation.

We have just a few slides, and then I'll have ann go through another part.

I'm larry weis, general manager.

So if our revenue requirement we have the test by function.

We $807 million.

The costs are listed at the bottom of the chart.

The production demand, and allocation of cost.

This issue is interesting in the standpoint that fundamentally, what it does, it creates an argument in classes of customers about which one works best for


me, and which one works best for others, there is, generally speaking, we end up with an argument about which one works.

That is the issue, frankly.

You will see what we have done is we have tried to recognize that, tried to come up with a method that is fairest to all the customers.

Going forward, keep that in mind.

We have the four coincidental peaks, recognize the peak capacity, the primary cost driver.

The other is the demand method.

The other is the energy waiting method.

The average, excess demand, assumes energy portfolio, designed to conserve peak in energy.

During 1994, when the rates were last worked on significantly by the utility, the grid operated differently.

That is that austin energy ran its own control area and dispatched its own for its own load.

That has changed in the ercot market, we have to keep that in mind.

The other is the time differentiated method, that is base load and intermediate peak.

If you want to go into that deeper, we can.

What we have done on this chart.

It is an important chart that we keep.

For every class, customer that we are proposing, the new rate design.

In each one of the lines, you can see a method that is used and what the bill impacts would be based on the different methods.

And the goal that we -- and our recommendation is aed, within the goal of an average monthly deal.

What we are doing is putting bandwidth with 5% above and 5% below to try and soften any impact of any of the methods to customer classes.


And again, we can go into that in more detail, if you like, and I will turn it over to ann at this point, for going to the next slide on this.

>> Ok.

As larry said, we --

>> hold on a second, ann.

>> So, larry, you mentioned that on the production demand allocation method, you could go into that more if we would like.

I would like to hear some background on them.

I don't know if now is the time and you would like to continue on with the slides.

>> We can certainly do it now.

Smor smor I know you said the different methods.

If we can understand the different methods, that would be helpful to us -- at least to me.

>> Let me step back to the slides.

There are three different types of allocation methods.

There are three different philosophies.

The demand responsibility.

The example that best fit austin energy is the peak.

That recognizes peak capacity as the primary cost driver.

The value is to serve peak.

It is allocated on the class contribution.

The system for system peaks.

One for june, july, august, september.

Those are averages.

It is the simplest of the methods.

before you go on,


don't we ever invest to serve our base load?

>> Yes, that is one of the faults with 4 cp.

here, the only reason is to serve peak?

>> Right.

Remember, there is a philosophy behind each of the methods, that is the philosophy behind it.

you weren't saying that is a fact, you are saying that is the assumption behind this?

>> That is correct.

>>Morrison: thank you.

>> The next is the energy weighting.

It has a different philosophy.

It assumes that both peak and energy have value.

So when you invest in your generation portfolio.

You do understand that it has value for peak and energy.

And so it is kind of a weighted method.

And it allocates the planned investment cost, based on the combination of demand and energy.

So it basically says that your average demand or your energy is equal to your system load factor.

The excess is allocated over the class peak.

The noncoincident peak, not the coincident peak.

It recognizes that there is a relationship between the class peak and your energy.

So the third one is time differentiated.

It is the base load intermediate peak or I will refer to it as bip.

It is much more complex than the other because it looks at every hour.

It allocates the plan investment cost on customer class loads and find a base load, intermediate, peak period.

Base load is allocated over energy.

Intermediate is over 12(c) p.

And peak is over 4 cp.

It is much more complex.


Assumes the dispatch.

Austin energy chose average and excess demand.

If you want, I can explain why we did that.

It is a little bit of a long story, but when we started the rate review back in fall of 2010, we were -- we hired the , formerly known as r.w.bakk.

One of the things they said is what production demand allocation do you typically use?

We told them, we have actually a resolution that dated back to 1997 that requires us to use probability of dispatch, we call it p.o.d.

So we would have to use that.

Well, we were in the deregulated market and it was in the nodal market was pending.

Nodal market was effective 2011.

With that in mind, they basically said and everyone agrees that the probability of dispatch would not align with that market, the current market we're in, because it assumes that you own your generation, you buy your generation to serve your load and that you dispatch it yourself.

Neither one of those are true in the nodal market.

one could assign, not probability of our own dispatch but assign cost factors associated with the market cost, couldn't we?

>> That is basically what the consultants thought.

is a time differentiated method, they said that would be similar to base load people, or bip.

>>Spelman: that is what it does.


>> You try to determine what kind of plant, if you have a peaker, it is for peaking.

If you have an intermediate load, or combined cycles, it would be for intermediate


production or generation.

, nuclear, coal plant, that would be base load.

You put those in that category and determine the dispatch on an hourly basis.

So because of the resolution we had and that was the requirement at the time, then that is why bip was selected for the public involvement committee meetings that we had in june.

But we also presented for coincident peak and average and excess demand for the same time.

We wanted to make sure we covered all the philosophies.

Then in the summer, after we gathered the input from the public involvement committee, we revised or updated, basically, our cost of service.

We looked at all of those decisions and we tried to align those with our strategic goals and rate-making principals.

As we did -- principles.

And we had experience in the nodal market, we realized, using bip you have to make a lot of assumptions, because you have to design a cost of service for each generation plant, then you put it into the right category and you dispatch it over an hourly basis.

So there is a whole lot of assumptions involved in bip that are not involved in average and excess.

We realize it really did not align for the nodal market.

In the nodal market, the peakers are not just used for peaking, we use them often for balance, for ancillary service, nothing in the bip model that would work with that.

That is one thing we were concerned with.

As we tried to align them with the strategic goal after and excess demand with the class peak aligned directly with our goal to save 800 megawatts of energy by 2020 through demand site management.

It looks at class peak and it rewards high load factor customers.

We felt it wass-- was evident it would align with the goals and thought it was better for the environment, utility and customers.

that statement you rewards high load factor customers, to simplify that -- I struggle with that a little bit.

To simplify it, it means they're using to me, it basically means they're using an even amount, a relatively uniform amount of energy.

And that doesn't necessarily mean they're being efficient.

That could always be using way too much and they would be rewarded here?

>> Let me step in here a little bit on this.

Load factor, if you have a large demand or small demand, it doesn't matter.

If you use that amount of demand around the clock, that is the highest load factor you can have.

You put a demand on there, to do that.

Residential customers, for example, have very poor load factors.

They have a big afternoon peak and not using energy the rest of the time.

So I am assuming you're asking for a distinction between those?

you are making a judgment.

You said poor load factors, they have low load factors.

They said it is poor because it is not an efficient use of our plant but it may be an efficient use of our energy.

I wanted to get that distinction.

I hear it rewards high load factor and high load factor is good because it encourages efficiency.

I don't think it does.

It encourages uniform use of energy around the clock and it could be way too much around the clock as opposed to being efficient and not too much around the clock.

To me, that is an important distinction to keep in mind.

>> I'm only looking at it from the perspective of the physics of the system providing economical dispatch to loads of customers.

The most efficient utility you have, you have a load that never changes, right?

The less efficient the system is, is having loads that change radically and often.

We're trying to be efficient in utilizing efficiency in energy.

That is the key here that we have to deal with.

As a council.

>> True.

>> I think one of the points to that, too, is that when you use the equipment efficiently, it lowers the cost of service.

That is how we apply it in this situation.

A high load factor customer would have a lower cost of service because you spread that equipment cost over more.

So that is kind of the way it is applied.

It is a little bit different concept.

>> And when you say that a.e.d.

Is better for the environment, i think I heard that, that is because it rewards the higher load factor customers.

Are you thinking -- is your reasoning that we're talking about reducing the peaks and that is where the environmental benefit is?

>> The reason I said that is because what we're trying to do is in the long run is help the customer reduce their bill.

or energy efficiency through pricing signals, that is how you accomplish that in rate making.

And that is why, if can you do that, you are helping the utility, the environment and the customer.

provides those pricing signals because?

>> Because it looks at the relationship between the class demand and the energy.

>> Right.

And generally, you want to incentivize reduction of the peaks.

So through efficiency measures can you bring the peak down and make yourself a higher load customer?

>> That's correct.

>> And that happens coincidentally with all customers.

>> Right, ok.


I think we interrupted you, so please continue.

>> I will just go on to the next slide.


As larry said with this slide, this shows the results of the production allocation methods.

And it shows you the average monthly bill, by class and by method.

So you can see the results of the three methods, just looking at residential, you can see the average and excess demand yields an average bill of $108.

And four c pvment -- 4 cp and bip.

If you look at primary greater 3 under 4 cp and 1.5 under bip.

It is just the reverse.

And the thing that when we looked at this, really, all three of the methods are acceptable methods.

And so what we decided to do was use it as bandwidth and as a test of reasonableness.

So we decided -- we also knew that the residential and small business were likely to have the highest increase in bills.

So we decided to deduct 5% from residential and 5% from the secondary voltage under 10.

That is the worship, portable buildings at school.

Those receive a very large increase because they have been -- they're further away from cost of service and some of the other classes.

So when you did that, you can look at the two blue columns in the center.

You can look at bip, you can look at the ae's recommendation.

You can see on the residential bill, there is only less than a dollar.

That is actually about a little over 50 cents for the residential increase.

So they're basically the same.

Small businesses gave them a reduction of $5 on the average bill.

You remember fairness is defined by cost of service.

We developed a bandwidth that was within the cost of service range.

We felt like this was a fairway to allocate it.

-- Fair way to allocate it, we felt like this was a reasonable test.

councilmember tovo, you look like there is something on your mind?

I want to get a sense of how much longer your presentation is, and at what point --

>> we can go through the next slides.

we'll hear from you and then comment.


>> This is a quick visual of the same thing.

If you look at the gray bars, that shows the highest method and lowest method, the blue represents the a.e.


For residential we align with the bip method.

let me interrupt you quickly.

Make sure we understand what the graph is.

That baseline is the current bill, the height or depth of the bill represents the change in the current average class, right?

>> This is done with the aec, but it is close.

on average, whether or bip, you're talking a 30 to 40% increase for customers that are far below cost of serve, no matter how you measure cost of service, they have to come up 20 to 30%, we can choose which of the bars, but it will have to be somewhere between 20 and 30%, no matter which we look at.

>> That is right.

this is for the current requirement and not what is the proposal on the table.

What is the difference?

>> 71 Million.

3/5 of the numbers, roughly.

>> Right.

>> This is another comparison.

I can go through it quickly or we can skip it.

It is up to y'all this gives you the major points we looked at.

We talked about like on the allocation perspective, this is , it looks at class coincident demand or your peak.

Bip looks at the generation supply.

It focuses on the hourly dispatch.

Hour and excess demand focuses on energy efficiency.

It promotes efficiency at the class level, basically.

The industry accepted acceptance.

4 Cp and ead are common.

Bip is limited, mostly on the west coast.

More complex and more assumptions.

Use in texas, the puc is used 4 cp, the most commonly used is aed in texas.

Bip, I'm not sure if it has ever been used very much at all.

is in less alignment, I would say.

Bip, when we prepared this, we thought it would support the nodal concept, but we're concerned about that now.

is a blend of costs, so it does apply, you can apply it to the nodal mark.

Cost shifting or cost causation, that is what we have been talking about before.

The 4 c.p. is always on peak.

Bip shifts more to high-load customers.

shifts more to low load factor customers.

And the last one is a.e.d.s.m.

And energy efficiency goals.

is more like that because it aligns with ercot.

aligns best of all with our goals.

councilmember morrison.

could you talk more about the comment on the nodal market applicability, ann?

You said your thoughts are changing.

Oh, look.

Wave, everybody.

[Laughter] so could you talk about that a little bit?

>> As I said earlier, we have a little experience.

In the nodal market when we first prepared this, it had not begun yet.

We're realizing when you prepare bip, you have to say, if I have a peaker, we will use that for peak.

If I have an intermediate unit that follows the load, we will always dispatch that to follow the load.

If I have a base load, it will always be run all the time.

In the nodal market, we're finding out because they dispatch based on least cost and since natural gas prices are so low, gas is even displacing goal.

So -- displacing coal.

Peakers are used to balance not, not just for peaking.

There is a lot of things that would be hard to argue if we had to defend this position.

There are so many assumptions built into it.

has anyone done a rate case since we went to the nodal market?

So we are on new ground here?

>> Yes, fully integrate.

Never been a generation.

Generation is competitive.

So they don't have to go to the p.u.c.

We will be the first one with something like that.

basically, what is getting used is defined by the market as opposed to by us.

>> In terms of generation.

in terms of generation.

Bip is actually more applicable.

Could you talk about why that is?

>> These were developed back when we were in the public involvement committee, I don't want to just change them because we have presented these all along.

I'm not saying you can't apply it to the nodal market.

It would just be more difficult.

Because, again, it is very complex.

Looks at every hour of dispatch for every unit you have.

So you can imagine when the market dispatches those, based on least cost, it is hard to predict that.

another way to look at it is we used to dispatch only with austin energy system, and needs.

We did internal economics.

Now we only dispatch to the ercot market and buy back through the ercot market.

So it is totally done differently how we use our generation.

I think I am a little confused.

Does this chart reflect your current thinking?

You said this --

>> the only one, if I prepared it, I might -- I would probably just take the nodal market out of there.

Because -- I think it would be difficult to assess that on all of them now.

Like I said, when we started this, we thought this would be great because we can just stack our units show how they're designed and dispatch them that way.

It doesn't work that way in the nodal market.

so you think that actually, there are big question marks about any of them for the nodal market.

>> I think it is difficult, the works better for us is because it is a blend.

The average demand is equal to the system load factor.

And anything above that is allocated based on the class peak.

So those things work with the nodal market, they don't contradict it.

we no longer have the simplicity of the internal market to service.

But we have the year's worth of information as to when we had to turn on the plants that we had to turn on.

It is there a way to use the historical, we know when we turn them on and turn them off as a basis per bip rather than our assumptions as to what is base load and intermediate?

>> The way that were dispatched is to dispatch to the market, not to meet our customer loads.

That is the disconnect that we have.

So we're looking at customer cost allocations and fall out and the way the market operates today, there is a disconnect between the two.

00 on wednesday afternoon, not only is it most likely we have to turn on our peak because the nodal market will make it make sense for us to do, that but our cost of energy whether we provide it ourselves, to ourselves or purchase it off the grid, it will still be higher, won't they?

>> It probably would, yes.

>> That is actually a good point.

That is what the demand production allocation method does.

We know it costs more to serve peak.

What you are trying to decide is who pays for the peak?

Which class pays for the peak?

the value of bip to me -- you could have a distinguished amount, arbitrary number of classes.

We are gauging, who is using this energy, less expensive energy, who is using that.

Whether it is internal, buying it off the grid, some combination of the two, it seems there is a fairly reasonable connection between our cost and the activities of our customers.

>> Well, really, it is more by the market.

I mean, today, the market price for peak can go as high as 4 megawatt per hour.

Our cost is not that much.

That might be driven, we don't know by which unit.

yeah, I think the difficulty is that every year will be different in terms of how --

>> in terms of how ercot region works and what we use to meet the necessary operation of the ercot market.

It is disconnected from what our customers are doing.

That is the hard part about this.

well, the marginal price of purchase power on the spot market is determined by what everybody else has going up by what our costs look like.

>> And how much generation available on the grid.

what all the other customers outside of austin, and customers are doing outside of austin, that is the disconnect?

>> Exactly.

>>Spelman: go ahead.

>> I can conclude with this one.

This is why we recommended a.e.d.

I think we probably already talked about all of those points.

I was going to ask our consumer advocate to come up and walk us through his thoughts on the mmatter.

ann, are you finished?



Come on up, paul.

Not sure your mic is live.

You might check it.

Hang on a second, while we fix that.

Go ahead.

>> Always go back to wires when you got them.


.. let's start with the way that austin energy has characterized the average excess demand.

They say it reflects energy and demand.

Well it doesn't reflect energy and demand, it reflects energy and excess.

When you add together the energy use which is the average demand, and your excess, which is everything above that, you get your demand.

They weight it so that if all the classes loads were coincident, the class peak loads were coincident, you would get the average in excess demand would be exactly the same as coincident peak.

This is something that I proved in a paper in 1980.

And everybody has known for a long time that average and excess demand does not reflect energy use.

The fact that you have to use noncoincident peak to have it even look like it doesn't do the same thing as coincident demand means that you have to use a demand measure that has nothing to do with how your production costs are incurred.

They're incurred based on total loads.

Austin energy's total loads, ercot's total loads which drive the market price for purchases, value of additional sales.

But they're total loads.

Not class loads.

The fact that street lights peak at night does not require you to run out and buy more nuclear power plants.

Essentially, what the method does is it allocates all of the fixed costs of the nuclear power plant and coal plants and the extra costs on the combined cycle, as if they were all combined by a few peak hours every year.

It is just not true.

It is not reality.

There is no concept of fairness under which you can say that average and excess demand is fair.

Fairness is a broad concept, it is what we are trying to do with cost allocation.

There are many situations in which you are left in a choice that would tax solomon to find some absolutely perfect way of doing a division of the cost.

But this is not one of the fine issues.

This is not a matter of we've got something that is difficult to allocate.

This is the reasonable way of doing it.

Average and incident demand, coincident peak demand, 12 coincident peaks.

Those are all completely preposterous ways to allocation.

Even peakers are dispatched a lot more than just peaking.

Just not your peak hours that it requires.

.. the presentation by austin energy has a lot of internal inconsistencies.

And it's confusing because they say two different things about the same allocation method, such as base and intermediate peak.

It is also because in their treatment of production cost allocation, they treat the nodal market as if it were something that fell out of the sky.

And up until a year or so ago, austin energy were an island that dealt only with itself.

And somehow if half of fayette was in austin, and the rest didn't exist.

And they just dispatched the piece that they owned, the rest of the world isn't there.

That is not the case.

Unless austin energy was extraordinarily incompetent, whenever it had fayette energy it didn't need to meet its own load, it dispatched them if anybody else needed it, because the coal was cheaper than gas, until very recently.

And the same thing is true for the gas-fired plants.

Now, with ercot taking over the dispatch, the decision about which plant will be warmed up and ready to go and dispatched at particular hours, it is probably more efficient, but this ieaking of a system that has basically been in place for decades.

It has never been the case that plants sat idle until austin energy's loads required them.

It is also not the case that you need capacity simply because of peak loads.

You need it because the amount of capacity you need is determined by both your loads and the resources that are unavailable, because transmission goes down, you can't get it to your service territory or to this part of the state or because a power plant goes offline.

So even for peaking capacity, in the simplest sense, even before you get into issues like following the changing loads, the changing generation patterns from wind, the peakers have always been required, for more than just a few peak hours during the year.

They may be required for many hours.

Or there are many hours in which they may be required.

Some years they're not, some years they are.

Another underlying confusion is this idea that somehow you're rewarding certain customers with a production cost allocation method.

Cost allocation is not rate designed.

The utility presentation basically said, well, we want to put a lot of costs on to residential customers because they contribute to peak.

And we think they're expensive to serve because we think , it is really peak, should drive the allocation of the fixed costs.

And that will encourage residential customers to conserve.

But then in the rate design, rather than putting the cost into the energy charges, which will cause customers to conserve, they take a large part and put it into a fixed cost that the customer can't avoid.

It is in rate design that you give incentives to switch time periods, to use less in the summer, to not worry so much in the summer, that that is a concern, to save energy, to switch in between time period, time of use rates, and so on.

Cost allocation simply determines the part of the total cost pie that gets put in front of each class.

How you bundle that into rates determines your incentives.

We're talking about fairness.

That is something ann said, we're trying to be fair in this process.

So for production cost allocation, for the allocation of fixed costs, you have to look at what's driving them.

And member spelman when you said, can't we look at the way the plants were dispatched in 2011, you could do that.

And you would wind up with something that is sort of a variant on the base intermediate and peak but avoids a lot of the problems in the base intermediate peak, which the company has used which is not the same as the way it is described in the handout.

There is a problem with using 2011 -- two problems.

One circumstances it was an extraordinarily hot summer.

Usually, for cost allocation, you want to use something like normal conditions.

I imagine this summer is not the same as 2011, that that is not typical.

To the extent that would change cost allocation among classes, whether you use 2011 or normal year, that is something that kind of bothers me.

pointed out is fuel prices are more peculiar this year than anytime I can remember.

And gas is backing out coal, which means the sand hill combined cycle is a base load plant this year, to a much greater extent than it would be in a normal year.

I don't know of anybody who thinks that the -- I forget exactly what the prices were for gas in 2011, but something like $3, that that price is going to continue over the next couple of years.

It my be cheap by historical standards, but it is probably not going to be backing out relative or modern efficient coal plants.

[One moment please for change in captioners]

>> I don't have that.

In trying to apply that right now in the middle of this case would be a little difficult if you actually want to get an order out before the end of the summer.

>> Spelman: We very much do.

I understand the value of doing something like that if we had a couple of year ramp up to it.

We could align our databases to produce that kind of information, but where we are we are and we definitely want to get done in a couple of months.

>> Right.

You are where you are, so i suggested something that i think is is fair and much simpler -- morgan chesky simple --

>> morrison: two things.

I assume this thing has never been done in the state of texas, yet?

>> I don't know what the approach we're talking about is yet.

>> Okay.

>> It's basically a probability of disappear based on -- on participation in the market as opposed to a hyphetica own load probability dispatch because austin energy has never dispatched as an isolated system.

They've dispatched as a part of a market where they were selling bilaterally on an hourly, daily, weekly basis.

And buying on that basis.

And a reality probability of dispatch would have reflected that whether you had done that in 1994 or today.

And today it's easier than then because you have information about how the system dispatched and what the prices are.

and I wanted to add that the pecan street folks are collecting volumes and volumes of data from the market.

We should put that in the back of our heads.

>> Spelman: For future reference.

Go ahead, paul.

>> Well, so what have proposed to do, and it's in the handout that I printed last friday for this session.

-- Distributed last friday for this session, I hope you have it -- let me see where I was.

Session 6.

I also distributed session 5, updated version.

So at the top of page three I have a table where I take the company's reported dollars per kilowatt year, the cost end rates, for each of their power plants, and calculate the part that's attributable to equivalent peaking capacity.

And by equivalent, I meant built at about the same time.

So if you go back to the previous page I have this graph that shows dollars per kilowatt year and the test year for each of the power plants.

And there are basically three clusters.

There's decker stem by itself back in the '70's.

In the mid to late '80's there's the decker, peaker and fayette, south texas, and then there's the sand hill peaker and combined cycle in the 2000s.

So what I said was if we attribute -- if we consider as part of the decker steam plant the cost of the same number of kilowatts for the decker peaker, even though the decker peaker came on several years later and it would have been deeper if it was built back in the mid '70's.

It would cost almost nothing by now.

But let's use the 1980's decker peaker for decker steam, for fayette which came on in a couple of years of the decker peaker and south texas.

And then for the sand hill combined cycle use the sand hill peaker, which when you average out over the various units also has about the same angelo and service dates.

So I did that analysis at the top of page three and if you take south texas' $443 a kilowatt and say $13 of that is -- could have been spent on peakers and gotten the same amount of capacity instead of the decker peaker as -- you built more decker peaker equivalent instead of south texas.

So about three percent of the cost of south texas can be considered peaker equivalent, peaker related.

And for fayette it's about 10%.

For the decker steam it's about 32%, which makes sense because you spend less on a gas steam plant than you do on a coal plant and less on a coal plant than on a nuclear plant.

And the sand hill combined cycle compared to the sand hill peakers is about 50% attributable to peakers and 50% to the energy saving investment in the combined cycle part of the plant.

And then I took an average, added up all the megawatts, added up all the dollars per cal watt, add up all the dollars and then calculate an average dollar per kilowatt, and it comes out to about 16% that is attributable to peaking capacity.

That is, if you had only built the system for peaking and built it at about the same time, you would have spent about 16% as much on your fixed costs.

Or that would be what would be reflected in rates now.

And the remainder of those fixed costs you incurred for the most part with good reason to reduce fuel bills because you don't want to pay peaking prices for all your energy.

Well, maybe you would have in 2011.

That would have been a good deal, but in general that would not be a good policy and I'm sure that it will be good to have non-peaking capacity in future years.

And in terms of allocation i say well, let's just allocate the 16% of production plant that can be -- that you would need just for peaking capacities to get the same amount of capacity, but just in peaking capacities.

Allocate that based on peak demand, coincident peak demand, and allocate the remaining 84% on energy.

You could get fancier and start breaking up the pool and this capitalized energy, the amount you've spent in fixed costs to reduce your fuel bill.

You could break that up into off-peak and shoulder hours and peak hours.

You could break it up between months.

You could do a lot of things along those lines, if you were so inclined, that would require more data and certainly mapping it to the customer classes would require more data than i have.

But right now what you do is you take fuel and take the gas costs, for example, and spread it out evenly across all the kilowatt hours you sell.

And that means during peak hours you're recovering more gas costs than you are in off-peak hours because there are more kilowatt hours being sold in the peak hours.

And you're recovering more of the nuclear costs in the peak hours than you are in the off peak hours.

Now, the way the company has illustrated the basinter mediate and peak method base intermediate and peak method on page eight of their handout, the -- it looks like for the base intermediate and peak, and people have done this sort of thing, certainly for probability of dispatch approach, you would say in -- the kilowatt hours consumed above about 7885 hours of use are all base load and every kilowatt hour used in that period would be charged a mix of coal and nuclear.

And the gas intermediate plants would be charged to the hours below the 7885 and the peaking plant costs would be charged for hours below about -- let's say about a thousand.

If you did that and you said we're going to charge usage in the very low load hours, exclusively for the expensive nuclear and coal fixed costs, you should probably also then try and differentiate the fuel costs, so if you're paying disproportionately for nuclear fuel -- for nuclear fixed costs, you would get the disproportionately cheaper fuel.

That gets complicated because then again you need the shape of the -- each customer class' load and you need to lay out all these hours of costs and figure out how these match up.

But it's one way of being fair and matching the fixes costs to the fuel costs.

My proposal follows what you're already doing with the fuel costs, which is that they're all thrown into one bin and take the energy costs and throw them into one bin, so more of the nuclear fixed costs are recovered on the peak hour than in the off-peak hours.

And some of the cheap intermediate capacity fixed costs are being recovered in those low load hours, not just expensive nuclear and coal.

And I think that fairly matches although as we've said before you can have a lot of definitions of fairness and a lot of ways of implementing it.

I think that's within the ball mark and it's pretty straightforward and simple and basically the calculations you see on -- in my handout are about all you need.

>> Spelman: Thank you, sir.

>> Of course I can go on --

>> Spelman: I can tell.

>> But if somebody would like to ask me a question.

>> Spelman: I'm almost certain you can go on.

You've made that real clear, but I bet we have some questions of you.

>> Morrison: Paul, I have a question over here.

So austin energy has also suggested the adjustment, they started with aed and then suggested the adjustment, 95% residential, up to 104% for some of the other classes as a mechanism to sort of even things out and perhaps make things fairer and all of that.

And they suggest then that on slide number 4 you can see the number shows that it's close, for instance, for residential to what the ip would have done.

Would have resulted in for the -- for an average monthly bill for residential.

So can you comment on that as a mechanism to smooth out and try to reach fairness.

>> I guess my question would be why would you want to use an allocation method that doesn't make sense and then say, but instead of doing that, we're going to -- instead of actually applying that, we're then going to soften the blow for a couple of classes.

And you can skip all this and just look at the total rate increase and say we want to be gentle to the industrial, so we'll give them half of the average rate increase and we think that residentials can pay a bit more, so we'll give them 125 percent and go through that way just based on what you think various classes can tolerate.

There's no reason to start with an aed allocation.

And if you use my allocation I think you probably wind up with, just on the production costs, I've got a graph on the -- of the percentage rate increases that would be indicated under each of the methods using the original 126-million-dollar rate request.

>> Morrison: Is there any way we can get that graph up for folks to see.

You don't have to have a drive that we can --

>> I can certainly send it to anybody who wants it.

>> Spelman: Or if you have a paper copy of it we could put it up on the projector too.

>> [Inaudible - no mic].

>> Spelman: That would be great.

>> Morrison: Because I was looking at this and I didn't quited in the legend.

>> Spelman: Which one is yours?

>> Well, so far all we've talked about is production cost allocation.

And austin energy has pointed out that nobody else raised any questions about allocation of other costs previously, so they didn't talk about it.

And that's a reasonable position.

And in my handout for today I go through a number of other cost categories and talk about where I think their method has been biased, particularly against residential customers.

So each of those is a -- each bar is the cumulative effect of a group of adjustments that I think would be appropriate.

The first one, the left-hand bar -- it's hard to see on this in the black and white version, but the left-hand bar is the company's original proposal based on the cost of service study, which they then offered later on to cut down somewhat.

And then the next bar over, which would be red, if you've got a color copy -- we've got it for each class.

The left-handmost group of bars is the residential class.

So we started at about a 25% increase being indicated for the residential class.

If you use my base and peaking approach, that drops down to about a 19% increase.

>> Spelman: That's again corrected.

>> That's again corrected, yes.

And I suggest changes in the transmission allocator, so it's the transmation allocation change.

I suggest a couple of changes in distribution so the gt and d, which is the fourth bar, purple if you have the color version, includes distribution.

And then there are the customer -- what they call customer of the costs, which are not the same as customer charges on the bill, but they're things like metering and billing, customer service, answering the phone when customers call and that sort of thing.

And I suggest a very crude adjustment to that.

I don't have the data on the individual cost components, but I make a fairly -- a simple adjustment and that gives you the last bar when you combine those.

>> And does the very first bar, the ae cost of service state of texas that include the 95% ajust.

>> No.

>> Morrison: It doesn't?

>> You take the 95% of that, which brings it down closer to the 20% line.

>> Morrison: Okay.

So the real question is -- okay.

So your point was that you didn't think that kicking off using aed was right in the first place, so why try and correct something that wasn't right in the first place?

So let's say we're going to accept something that wasn't right in the first place there your point of view, and then correct it so that we're about where we were if we had taken this approach -- if we had taken this approach, some of these other approaches.

The question is concept actually why not do that?

Is that going to be a problem later on down the line?

Is it going to be an issue for hitting some individuals worse than others or -- that's what I'm sort of trying to get at?

Because I think we had some -- we did have some concerns raised about that if you look at ae's table -- I guess it's page three of the handout that shows the difference between their adjusted aed allocator and bip, which comes in closer to what I would propose, you see that there are classes that are shown as deserving significantly higher rates than they would pay with bip and others that would pay less.

And your choice about which classes you moderate the aed effect for affects what happens to all those other classes.

It's not just residential and secondary, but also by my -- my analysis is that the medium and large secondary customers and the small primary customers should all get a lower increase than the system as a whole, and that the -- the larger primary customers should get higher increases.

And it's -- for all the classes other than residential and small secondary, there's no adjustment to undo the problem with the aed.

So from the residential viewpoint, it's not a problem in terms of equity for the other classes -- for the other classes it's more of a a problem.

>> Morrison: I know it definitely hits some of the other classes, and that's where our religious institutions and things like that are having trouble.

>> Spelman: turnek, you've been -- presumably you've recommended this method or some method close to this in the past.

Have you ever recommended this method for a utility based in texas?

>> Well, I did in 1980.

[ Laughter ] and I was not successful.

>> Spelman: Okay.

Where is a method like yours in use right now?

>> Well, let's see.

In novia zero scotia, some of the fixed costs are allocated on energy.

In man toby bah it's 100%.


In the six states that are by pacific core, utah through wyoming and oregon, washington, it's 25% energy.

There are a number of northwestern jurisdictions that use primarily energy.

>> Spelman: Okay.

What would --

>> oh, and a number of jurisdictions use methods like this, usually called something something probability of dispatch, but many of those have since restructured and aren't in the business of allocating generation costs anymore.

Massachusetts would be an example.

>> Spelman: Okay.

So your primary objection, just so I can understand, in a slightly broader scale, is adp and acp are both allocating costs entirely on demand, on peaking, and not on energy use.

>> Right.

>> Spelman: And a more fair allocation from your point of view would be some combination of peaking and energy, which aed theoretically does, but in fact fails to do.

And then some podlike method, which you've proposed one, would be a better allocation of peaking and peaking costs and energy costs.

>> Yes.

>> Spelman: Okay.

And roughly, what combination of peaking and energy cost does your method come up with?

It's about, what, 55?

>> About 60% on peak -- 16% on peak, 84% on energy.

And you could do other things I'm sure that would have different splits and different allocations based on different sort of theories about what would be a fair way of dividing things up.

>> Spelman: I'm virtually certain we have more questions for you, but i think I'd like to hear from weiss first, if we could.

Larry, you've heard all this and I presume you've heard all of this in the past.

I wonder if you could respond to it.

>> Well, I think it illustrates that it is a very philosophical issue and that -- here's one thing that's unique and I'll try not to get too technical.

Austin energy does not run its own control area anymore.

It used to.

Ercot is -- there's nine regions of electric grids in north america.

Ercot is the only grid in north america where the utilities within it do not run their own control area.

It's run by ercot.

We used to have probability of dispatch until we had a resolution here before you where we rescinded that, if you recall, and the reason for that is because we do not dispatch for our own load anymore.

We really basically dispatch for economics and availability, and because ercot asks us for the resource and we get compensated for it, therefore our wholesale contribution, which has not been very robust frankly ever since the mid, early 2000's.

So in other words, I guess what I could say is that we could have an economist or anybody in austin energy, we have our consultants who could have a counter point of view to a lot of that.

So I'm not disagreeing with what I heard, I just know from experience that there are many points of view that we could come at from this at and I appreciate the difficulty, though, of -- for you to ultimately decide what is there.

I can say that there historically has been a large cost shift in existing austin energy rates between commercial and the residential.

And that public utilities across the country a lot typically have that.

It's one that's kind of developed out of policy.

It kind of falls out from doing this.

So I'm going to turn -- give anne -- do you have any comments that you want to make?


>> well, I would like to say that the aed method that we use does consider energy.

In fact, the average demand is the same as energy in our calculation.

And that equals our load factor, which is 57%.

So about 57% is allocated on energy.

And the excess demand is then allocated on the class peak.

So it does weight those effectively.

And the method -- I just would like to say too the probability of dispatch method was looked at by about four consultants before we started this.

And no one feels like it would be acceptable at the public utility commission.

And aed is the most prominent allocation method in texas.

>> Spelman: Let me ask you more about that because since we are the test case, we have certain constraints on our behavior.

What methods are -- have been in the past acceptable to the puct?

>> Typically for cp and aed.

I'm not the expert.

I'm sure there are others.

They just -- texas is not a marginal cost state, and the proposal that we heard is typical to a marginal cost theory, so it just -- and i think in 1980s when this was presented that's why it was rejected.

>> Spelman: I can't help-- the economist in me cannot help but leap on that.

Of course we are a marginal cost state.

All states are a marginal cost state, it's just that texas doesn't recognize that it's a marginal cost state.

>> That's correct.

>> Spelman: We're not acknowledging that particular economic truth, and there are only two methods which are in common use that are acknowledged by the puc.

>> There's probably more.

Bob bailey daily is our utility regulatory person and let me ask him if he would like to comment.

>> Those are the two common.

Cp, 4 cp and some variation of a and e.

chernek argues that -- I bet he would be fairly persuasive that aed and 4 cp are basically both 100% peak or very close to it.

And my reasoning for agreeing him in advance of seeing his math is that our math, which shows the allocation for 4 cp and our allocation for aed are almost identical allocations.

So it seems to me that what we're trying to do is come up with a mix of peak and energy that somehow our mix of peak and energy has turned out to be identical to 100% peak.

And maybe that cast some doubt whether we're accomplishing with aed what we think we're accomplishing.

Would you like to respond to that?

I'm trying to summarize chernek's argument as i understand t.

>> I think the aed is really classified as an energy and demand method, not just a peaking.

chernek peak or split is also an energy weighting method and they're described in the same section.

To the aed really does recognize both peak and energy use.

>> Spelman: Is there any other method, sir, that the puc recognizes or has recognized even if not they don't encounter it frequently, as a mix of energy and peak?

>> There can be variations on the a.

So our proposal uses an aed with an ncp to allocate the excess demand.

You could use a different allocator, like 4 cp for instance has been used.

>> Spelman: Okay.

And the ncp means basically that class' peak as opposed to the coincident peak of all classes added together.

>> Correct.

>> Spelman: Okay.

Is there any reason -- is aed ncp a standard by itself or is there different ways of approaching aed?

>> The only variations I'm aware of is how you treat the excess.

The average is just energy, average demand.

And then the excess piece has to be allocated in some way.

We've allocated it on class ncp, but I've seen where you allocate the excess on class 4 cp.

>> Spelman: We could use 12, we could --

>> yes.

>> Spelman: Gotcha.

Other questions?

>> I just had a couple of factual corrections.

The power pools, the integrated, the to's around the country, wherever they exist, new england, new york, the mid atlantic and pjm, they're control areas.

There is no control area in new england other than iso new england.

So it's not true that texas is the only place where that exists.

And the probability of dispatch method was used by many jurisdictions that used embedded cost a allocations.

You don't have to use embedded.

You can use probability of dispatch.

Probability of dispatch can be used to allocate the average embedded accounting costs of your plants, the revenue requirements based on what exists today without any reference to marginal cost.

So that was also a mistake.

>> Spelman: If we were a marginal cost state, that was the primary factor, it would be a lot easier to justify pod, it would be more difficult to justify other things.

>> I don't see the connection because if you say we want to charge the people who use energy primarily in off peak hours for the fuel and the plants that serve them in those off peak hours, and we just take a slice off that, including plants that are not marginal, including the nuclear plant, for example, which is never at the margin.

It's not providing marginal fuel, it's not providing marginal capacity.

It's just there.

As a matter of fact, probably can't turn it down if you wanted to under most circumstances.

You include that cost.

That's an embedded cost, average cost approach.

It's not a marginal cost approach.

Marginal cost approach is just look at the cost in the most expensive unit.

Now, you can do that with a nodal price system, a regional generation system if you want.

One other thing I wanted to point out is there's there's some semantic argument that because you use average demand, which is the same as energy, in the average and excess, therefore it's weighting energy along with demand.

But it's not weighting energy and demand, it's weighting energy and excess.

And the excess plus the average is your demand.

And then you weight it in such a way with using the load factor so the math works out so it's coincident peak if all the classes peaks were happening at the same time.

Some jurisdictions looked at average and excess and said that doesn't make sense because it is collapsing to peak demand.

We like the idea of allocating energy and they did something very simple and just said let's use average and demand and allocated some percentage in the case of austin it might be 57% on energy and the remaining 43% on peak demand, not on excess, but on the entire peak.

That would be doing what the average and excess is often described as doing.

>> A nice, simple brute force.

We need to figure out what figure is energy, what is demand and I presume we would have the data at hand to be able to do that.

>> As long as you keep it in that kind of simple question, yes.

>> Spelman: And if in fact your mathematical proof is wrong and average and excess demand does exactly what it's supposed to do, we should come up with the same answer either way, so there wouldn't be any harm in doing it that way.

Is that something we can do or is it more complicated than it appears to me conceptionally?

>> You can develop an average and demand allocator.

>> Spelman: Do you suspect that the public utility commission would find that persuasive at all?

>> I don't know if that approach has been accepted or recommended at the commission.

>> Is that going to be highly resource intensive from your point of view?


Questions, comments?

Go ahead.

>> Just a quick one.

It sounds like that particular configuration hasn't been presented to the puc before, but it does sound like there have been various alternatives -- alternate versions of aed presented, so it's not uncommon for a utility to come before the puc with a tweaked version of aed.

So in in fact that would be consistent with what they've seen before.

>> Okay.


>> Spelman: Other comments?

Councilmember morrison.

>> Morrison: Because we're first out of the block here going -- doing a rate case with the nodal market, i think that it probably makes sense to really try to explore these different kinds of options to do that.

And I don't know if we need a motion to get this going, but I would like to look at that if it's -- since it's not too resource intensive and see if we can get that alternative in front of us.

So I think that would be a motion or a direction to staff.

>> Spelman: It's my understanding that staff would prefer the form of a motion.

It's clearer that way, is that accurate?

Either way, whether I -- i suspect we all have the same feelings on this, but let's clarify what our direction is.

Why don't you go ahead.

>> Morrison: [ Inaudible ].

>> It's up for discussion today and not for action.

That was in accordance with the schedule that was adopted.

>> Spelman: That's right.

>> Spelman: Let's do this informally.

Would you like me to take a crack at it?

Since our interest is in producing a rate structure which includes both energy costs and demand or peak costs, and a simple way of doing that relies on information we already have available as to the total energy usage by class and the peak energy usage by class, for example, the 4 cp stuff, then a simple way of accomplishing what we've been trying to accomplish with aed would be a weighted average of the energy cost and the peak cost.

So what I guess we're asking to you do is come up with the weighted average allocated among our -- the customer classes that you've already identified.

And let's take a look and see what the average bill or the total cost or whatever marker you want to use would be among each of those customer classes if we did it that way.

Seems to me what's missing is the proper weight between energy and demand.

I wonder if you guys could give me some guidance as to what you think seems like a tolerable waiting scheme.

>> I think we could have a recommendation on that.

>> Spelman: Try a bunch of them.

>> I think we could try a bandwidth, could we not, bob?

>> Spelman: A baseline, it seemed to me, and I don't understand to recommend it because it's so -- I don't know enough about this to snow know how seriously to take this.

But you've got approximately half of your production cost allocated to fuel and purchase power, which is clearly energy.

And about half of it too o and m, debt service and capital, which is pretty clearly demand.

You could probably argue square or flat that a 50/50 allocation seems not an unreasonable place to start.

And if you could come up with a better set of allocations, I think you would be perfectly free and that would be with the direction of the council.

Is that -- does that sound reasonable to you guys?

>> Morrison: And the results will be in the form of -- the results will be in the form of an assessment in the change of the allocation from the current -- between the different classes?

>> I think we could show you a range of options similar to slide three.

With some different allocations used.

>> Morrison: And I know sometimes we talk about a change actually shift some of the costs from one class to another.

So if we could track that also, I think that would be helpful.

>> Spelman: Good.


Anything further we need to say about cost allocation, mr. weiss?

>> So -- well, part of the procedural part of this is that as we understand that one of our coming up meetings we've got a range that the decision point that we would go through again and that would be the time that we would presume that possibly a decision on that.

>> Spelman: I think you could reasonably anticipate a decision at whatever meeting we have scheduled for making decisions that is next.

But you should try and get whatever results you can to us as quickly as possible to make sure we have the maximum time available to think about it.

>> Yeah.

We'll make that all public and everything.

>> Spelman: Okay.


Let's change gears.

That's item 1 is concluded.

Now we can go to item two on our agenda, which is to talk about the schedule going forward.

Are there any -- let's remind ourselves of where we are.

This is session -- good golly, where are we?

This is session 6, we've just done that.

And the next session we have scheduled is for the 23rd on thursday from 10 to 1:00.

>> Tovo: I'm afraid we have a few.

Councilmember spelman, if i could speak.

>> Spelman: Please.

>> Tovo: It looks to me that session 6 right now is listed as tuesday, april 19th and of course today is thursday.

BUT -- THE 21st.

>> Spelman: Today is the 19th.

>> Tovo: So we have a few glitches on the revised schedule we have before us.

But I think that the main changes that we need to consider is session 8, which is now scheduled for april 24th, you notice in blue highlights out of city ratepayers has been added.

It was there originally.

We removed it because we were having the session in lakeway, but we haven't had a chance to discussion what should be done about out of city ratepayers with regard to rate design.

So that would provide us with an opportunity to talk about that on april 24th.

And then we do have.

We do need to schedule extra sessions so those have been added at the bottom.

We have a few challenges with today's.

As you know we didn't begin talking about disconnection and connection, which was scheduled for last week.

So we need to make that up next week.

Am I right in thinking that we had a monday or tuesday work session scheduled for next week?

>> Spelman: Monday the third is actually a date which is live.


The next date we are scheduled to get together on ae issues is monday, 00 in these chambers.

Is that accurate?

>> So on that date -- i would like to talk about the room as well, but on monday we'll also need to pick up disconnection and connection fees, off system sales, the review -- basically all the other -- basically all the to be scheduled items at the bottom which were scheduled to be talked about at the last meeting and we ran out of time.

So on monday we'll be talking about other related issues, I propose, reviewing the audited finances, talking about off system sales, talking about disconnection, connection fees and changes.

And then we should keep on there the potential preliminary council action.

I would also propose that we --

>> Spelman: This is all FOR MONDAY THE 23rd?

>> Tovo: It is clearly too much.

I think we'll have to figure out, depending on how far we get, we'll have to shift what is currently contemplated for session 7 probably to session 8.

>> Spelman: Okay.

So you would not be talking about -- you're proposing we not talk about any of the contents of session 7 and the next session on monday just be a makeup session to fill in all the gaps that we've left so far?

>> I wasn't necessarily proposing that, but it may be the most progressmatic approach.

I think it very unlikely we will get through all of our last session and potentially take some preliminary actions and move on to session 7.

What do you think?

Or colleagues, what do you think?

Is it realistic?

Basically we will be merging two and a half sessions into one.

>> Why are we merging two and a half?

>> Spelman: I understand one and a half.

I don't understand two and a half.

>> Tovo: We would be doing session 7 topics.

We would be doing the items that are listed as to be scheduled for almost the entire session five.

And we also haven't made preliminary actions on things, which was about half of our meeting on session 4.

And it makes sense, you know.

One and a half, two and a half.

I think the basic point is we've got some catch-up work to do.

>> We're going to need to schedule extra sessions.

>> Spelman: What is the hurry in picking up the to be scheduled items?

Is this the right place for us to take a look at these?

>> Tovo: It is because they're revenue requirement items.

They're the last bit of issues before the revenue requirement before we move on to cost of service, which we did today.

>> Spelman: Okay.

And the stuff that's picked up from last time of course is revenue requirements.

That's where we were.


So your proposal is --

>> my proposal is that we move the to be scheduled items on to monday's agenda, MONDAY APRIL 23rd.

That we there potential preliminary council action from previous sessions.

I think in the interest of keeping it as general as possible from preliminary sessions.

Or from earlier sessions.

>> Morrison: One idea i had heard was that that we so that we can move things around and we could then -- I think our legal department suggested this.

So we could focus on the things in order, but we don't have to continually run into trouble with posting items.

Would that -- something to that effect?

>> Tovo: If if that's a legal option I think we should do that because we are running into -- unfortunately today we came today prepared to discuss disconnection fees and we couldn't and we couldn't take preliminary action because it's not an item.

I think that's a great approach.

>> I think it would be potential for from a posting item because staff needs to have a bit more detail understanding as to what would be --

>> Morrison: The order and the priorities.

And it might be helpful if we could find some way to identify what the priority topics would be for a particular session, but it will just allow it the flexibility so we can make the best use of our time.

>> Tovo: So because we're not posted for preliminary action today --

>> you are listed for action on the schedule.

>> Tovo: Okay.

Then I move that we do what has just been discussed.

>> Spelman: Whoever is actually putting the schedule together, do they understand where we are?

>> Morrison: Shirley, i think the clerk is -- is the master of the schedule, is that correct?

>> So let's talk about how the rooms are booked for next week.

We have you on the schedule for monday morning.

And for tuesday afternoon to talk about ae, in addition to your tuesday work session.

So what I'm hearing you say is you're not changing any of that, and that what you want from now on is just a list, all the remaining topics, and somebody is going to decide which one of those topics get on monday's list and which one gets on tuesday's list.

>> Spelman: I think it's probably cleaner than that.

Kathy, do you want to try again.

>> Tovo: I will.

What we want to do is post the entire schedule from here on out, all of the topics, because we need to go back and pick up some of the previous topics.

We can identify priority topics for each time and we've got those pretty clearly laid out.

reese, you look like you've got a question.

>> I was going to say something.

It's spinning in my mind developing all this and making sure we have the resources, so I have a suggestion.

We've come to these meetings every time, we have a little presentation and those take time.

And maybe they're not really necessary as much each time.

That's a question.

In other words, if we get into question and answers, the faster we get into questions and answers, the faster we'll get to the issues.

So if we have to spend a lot of time -- we don't mind making the presentations, but I'm looking for some direction from you on whether those are valuable because they're really -- they're really information that we've already presented many tiles in different forms.

And it's just really a question and it's more of a question of efficiency and what you would like in these presentations.

>> Spelman: I think your presentation today was a good example for what would be appropriate.

Your presentation itself only lasted about 10 minutes and the vast majority of the time we were spending in discussion.

But I think refreshing our mind as to what the primary issues are on a topic is time well spent.

>> So we could bring kind after high level on each one of the topics and something kind of high level.

And as the questions get deeper we can draw on our resources to bring it out, i think that would probably be a better way because when i hear anything might come up, then -- but I understand, i understand that that might be the most efficient way to do it.

I just want you to be cognizant that we're not going to prepare much and we'll just have the high level stuff.

>> Spelman: I think high level stuff and little preparation is an appropriate use of your time.

>> Morrison: I was going to say, and in fact -- i think we're still planning to go in this order, we're just trying to add the flexibility so we can pick up the items that we didn't get to and you all have already prepared -- you've prepared some stuff for us already off of some sales and disconnection fees.

And perhaps you've already prepared the charts.

I'm not sure.

>> Councilmember, would it make more sense then if i post the agendas more in a retrospective sense, that is the topics that were discussed would remain on the agenda, but that we not post all the way into future?

That way you're retaining the freedom to discuss topics that have been discussed or maybe were missed, but you're not necessarily posting to topics that will be discussed three weeks from now.

>> If there's a legal way of doing that, that would be great.

>> Yes, there is.

>> Spelman: Okay.

Are we all clear as to what we'll be talking about for the next meeting?

>> Morrison: Let me repeat what my understanding is.

My understanding is that we are going to discuss -- that we are going to -- the items that are in blue at the bottom, other related review of audit finances, although i understand did we already hear that or not, the 2011 audit?

>> No, we did not get to that.

>> Morrison: Okay.

So these four items, other related issues, 2011 audited finances, off system sales in a nodal market and connection or disconnection fees and charges.

So we want to pick those up.

>> Yes.

>> Morrison: And then at this point make those an addition to item number -- to session number 7?

Is that the expectation?

>> Yeah.

I think we need to hit those issues before we move on to issues underneath session 7.

>> Morrison: And it's not realistic for us to get to those and everything that's already listed in session 7, but if we just insert them, then we'll get as far as we can in those items.

And perhaps begin a discussion of other rate designs.

>> Spelman: Okay.

Everyone clear?

>> Tovo: Councilmember spelman, if I may.

>> Spelman: Go ahead.

>> Tovo: We just really need to keep on there potential preliminary council action issues from the earlier sessions in case we do feel moved to make some preliminary decisions with regard to the transfer rate or a couple of the other big areas that we haven't yet taken any preliminary action on.

So as long as that remains i think we're good.

>> Yeah.

If you were thinking five together you would be keeping potential preliminary action for sessions one through four and adding five to six for that.

>> Spelman: Fair enough.

That's a motion, but I will not ask you to restate it.

Is there a second?

All in favor say aye?



By a vote of four to zero we've changed our schedule.

That's all we're posted to do.

>> Morrison: If I may, i think there is -- we were meeting -- we moved into this room because they're long sessions and it makes a lot more sense for the folks in the audience to be in this room than in the boards and commissions room.

As a work session we had originally started by working at tables around in front.

So it was more of a back and forth, which I think is frankly really important because it's about us working together with everybody as opposed to just back and forth with staff.

I would like to figure out how we can go back to that configuration.

And I don't know who answers that question.

>> Spelman: Rudy answers that question and the answer is yes.


>> [Inaudible - no mic].

>> Spelman: Councilmember tovo.

>> Tovo: If I may, I was told that it requires an affirmative vote of four councilmembers to move back to the floor.

So I don't know if -- again, we're not posted for preliminary -- it is related to the schedule.

So I would move that we move back to the floor.

>> Spelman: That's a motion.

Is there a second.

>> Tovo: For all the reasons councilmember morrison indicated.

>> Spelman: Seconded by councilmember riley.


All right.

I need a second.

You want to object to this, councilmember riley?

>> Riley: [Inaudible - no mic].

>> Spelman: Seconded.

All in favor say aye?



We've got the four votes we need in order to move back to the floor.

Is there anything else to do with the schedule?

>> Tovo: I do want to --

>> Spelman: Go ahead, councilmember tovo.

>> Tovo: I want to raise a concern that mayor pro tem cole indicated to me and i know she brought it up in the last 21 and that was the relationship between the budget sessions and our work sessions.

And I raise it without any clear solution, but I hope that maybe when she's back next week we can take up the issue again.

I think there was -- I think she was interested in starting a discussion about whether or not we should consider as a council moving some of the budget sessions.

The other alternative of course to be to stretch out our work sessions for austin energy.

So maybe we can post that as something to consider and then take it up again next week when we have a full council back.

>> Spelman: That sounds like a reasonable thing to do.

At some point we might also consider whether or not we need to cover the rest of the ae issues in the same level of detail that we've been covering them or whether or not there's a possibility of putting some of those projected sessions together and reducing the size of our schedule, especially since we are one session late as it is.

But that's another thing which is on the table for next time.

There being no further business to come before this COUNCIL, WE ARE ADJOURNED