[BLANK_AUDIO]. Okay, so, Don you've talked about the economies of scale and the data here what about for other components Great question. Right now I'm thinking of two kind of classic studies that have been made of the costs of different water system components. Pipes water intakes wells elevated storage tanks treatment plants pumping stations that sort of stuff. and, probably the most reliable data that I know of and I've been away from this business a little while was a study that was conducted by the US Environmental Protection Agency. It's in figure four. >> So we can give the reference to the students. Yeah, okay. >> And in that study by the US EPA it was a gigantic study it had thousands and thousands of data points for different components each component with the systems. So pretty good reliability. All of the, all of the cost equations for predicting costs are the same log linear models with cost on the left hand side and some capacity indicator on the right hand side. And in component after component irrespective of the kind of component the exponent of the hydraulic capacity of this system is a number between zero and one and it's usually in the ball park of 0.6 or 0.7. Chemical engineers have done a lot of work in this field with cost functions for different manufacturing facilities. And they talk about the 0.7 rule being the exponent of the capacity of the system has a numerical value of about 0.7. And we know that if the exponent of these log linear functions these power functions if the exponent of the variable that reflects ca, capacity of the system is between zero and one the function is concave. It reflects economies of scale because the average cost decreases as the capacity or scale of the system increases. And it's remarkable in the EPA report, report that the exponents are in the order of zero point. It they can get to be as low as 0.5 that would reflect very large economies of scale. Next point of zero would be a, a perfectly flat fixed charge function. Tremendous economy scale. So 0.5 to about 0.7. >> So, do we ever have diseconomies of scale in, in in water in sewer systems? >> Let's think about. Let's go back to the water systems they're easier to deal with. There is alway, there are always economies of scale with respect to the hydraulic carrying capacity. That's just the nature of pipe. I said to you a six inch diameter pipe has twice the flow carrying capacity as a four inch diameter pipe. So with respect to the, the flow carrying ca, hydraulic capacity of the system there are always economies at scale. How about we're talking about water now how about the length of pipe? Are there economies of scale or just economies of scale? Let's think, you and I have worked in a bunch of countries. Right now I, I'm thinking of, of Ghana there in Kumasi they were, most of the, most of the households were in buildings that had four, four or five stories to them maybe six stories to them. It doesn't, in that, that kind of a community it doesn't take any more length of pipe to bring water to let's say 20 or 50 households that are in the same building as it would to serve a single household. So there are large economies of scale with respect to the length of pipe where we have multi-story buildings for housing purposes. Lets think about now a situation where we have a geographical area that the network is going to service. Still thinking in terms of a water system. But now maybe the mayor of the town maybe he lives ten kilometers out of town. And now this guy is saying hey, I want to be served with this system. Hm. Oh. Okay. So, to bring a very, very small additional quantity of water to serve the mayor large economies of scale with respect hydraulic capacity. But to extend the pipe ten kilometers to serve one household I'm sure that the cost of that extension the marginal cost is very, very high. That's going to pull up the average cost there are diseconomies of scale. So, and the ways this plays out especially with sewers you know, is that as we move away from the more densely populated parts of the city where there are economies of scale with respect to both hydraulic capacity and the length of pipe that goes into the system. Because households are close together or there are multi-story buildings. As we move to the fringes [UNKNOWN] where the population densities are low then there can be diseconomies of scale with respect to the length of pipe and that's where networks typically stop. And engineers may or may not have some thumb rules that they use for how far should we extend the pipe network into rural areas. >> Can you explain what you mean by thumb rules first? [LAUGH]. >> Yeah, okay. They may, they may say if the households are more than 30 or 50 meters apart from each other the length of pipe to serve an additional house pretty long. And the average costs of serving households in such a sparsely dense, sparsely populated area are going to be very, very high. The thumb rule would be don't extend the pipe network to areas or neighborhoods where housing does, the distance separating the households is any more than 30 or 50. This gets played out especially in industrialized countries with respect to when do you switch from septic tanks to a pipe sewer system? Septic tanks are on site wastewater treatment systems sewer disposable systems. So their costs are independent of housing density. Everybody has their own separate system. But septic tanks after a while are going to fail. What happens in cities the drainage fields that fail with these water, sewage starts to pond on top of the ground, kids play in it, it becomes a threat to health. And that's when usually the public sector comes in and says we're going to build pipe sewers. But what happens in those area if the housing density is still very sparse? Big time diseconomies of scale to serve them. That sometimes happens not so often in developing countries because it's, it's really just too, too costly. So, developing countries frequently will say okay, let the kids, we're going to turn a blind eye to the health implications here let the kids play, play games in the, in, in sewage that's on top of the ground which is very deplorable. >> Yeah. >> Just heart wrenching. >> Don you remember our students are most of 'em are not engineers so, you know, they're not going to know all, about all the details of the cost functions but I want to come back to some big questions about sort of cost and serving households in developing countries. First sort of how different are the cost in industrialized countries say New York or London from cities in developing countries Mumbai or Tegucigalpa or [UNKNOWN]? You know, what, what kind of factors determine the differences in cost between industrialized and developing countries? >> Well, the thought that immediately comes to mind Dale is that in the industrial, industrialized countries lots of rules and regulations. Lots of permits that need to be obtained before a contractor can come on the scene and build anything. Lots of rules and regulations in the bidding process which means that the bids that contractors in industrialized countries submit they're going to be more expensive than in developing countries just by virtue of the regulations that exist in the industrialized countries. But beyond that I think that developing countries even in large cities like Delhi like Metro Manila big places where you and I have worked that it's, it's easier to, for contractors to do business by getting their equipment on to the site in, in a place like New York City where for example there are more, there may be more vehicles. So, traffic control or the construction of temporary roads while we're tearing up a main highway to lay down a water or a sewer system on the edge of that highway and a right of way. All that traffic in an industrialized country has to be has to be rerouted someplace else. A few years ago Dale the value of alpha in in the pipe cost function was somewhere around maybe $15 US if diameter of pipe was measured in inches. >> What does that mean? Let's go back to that function we were talking about. It is cost per unit is equal to diameter alpha beta let's assume the value of diameter is one. One raised to anything is still one. So alpha what it stands for is the cost per unit length of a piece of pipe that has diameter one inch assuming that we've measured diameter of the explanatory variable in inches alright? That's around 13, $15. It's probably a good deal lower than that. Maybe no. It's probably a mistake for me to say good deal lower than that. >> Mm-hm. Somewhat. >> Certainly it's going to be lower than that in rural areas. So I wouldn't be surprised in rural areas or it's not really rural because rural areas are not going to have a pipe water supply system. But in a small town for example I wouldn't be surprised if they could find numbers that would be under ten okay? Maybe between five and ten if they had measured diameter. >> Now you've introduced this notion about urban and, and rural. What can you say about sort of the differences in cost in rural areas and urban areas in developing countries? >> Oh, in developing countries the, the the rural construction costs in rural areas in general are going to be much lower than in the urban areas. So I don't have any numbers Dale that I can toss out. But this is, this is, here's a consideration for I think, people for maybe your students who will be working in rural areas in developing countries with, with mostly piped water supply systems not so much piped sewage systems. In a country like the Philippines small town after small town are what I call linear systems. They develop along a major highway and they're laid out along a major highway. So if they're going to have a pipe system whether pipe water supply or pipe sewage if it's a pretty big town they need a lot of pipe. Because the, the community may only go one road back from the main road. They are, the word and, that engineers use for these pipe networks is they're not reticulated. Reticulated being interconnected pipes that are in close proximity to one another that surround blocks were, were people live. So in, there are a number of, and in my experience a number of communities especially in places were I've worked in Southeast Asia for example but you'll find it in Africa as well these linear communities. Sometimes their cost will be out of kilter with the costs of that one finds in industrialized countries because they’re linear systems laid out for long distances along roads. There are lots of vendors that are selling, doing business along the roads. You don't find that in, in, in the industrialized countries. So, and even in developing countries some small towns can be very compact you're not going to find linear systems in mountainous areas for example right? So, it's really hard to generalize on this stuff Dale. >> Right. >> And it's why we started this conversation by talking about the nature of the systems that we're going to model. So people have to think about if they're actually going to submit a construction bid on the system how would those construction bids change from one kind of system to another? If they would find it very difficult to think about how can I estimate the cost of this for submitting a bid chances are not, it's not going to be very easy to model that system. >> Most of our students are not going to be presenting bids though but I would like them to know sort of something about the cost per cubic meter of piped water systems and maybe pipe you know, cost per cubic meter for water and sewer. So, you know, when they're looking at cost data I mean, and they see you know, X per cubic meter for serving a community in a developing country what should surprise them and what X would not surprise you? I mean, you know. >> These are great questions and we're you're asking questions you're asking me to generalize. I'm not, I'm never very good about general. My head just doesn't work in terms of general, generalities. But I think, after decades of doing this kind of work both you and I pretty much agree that a well run water supply system these days the entire cost of of operating that system and this includes the cost of recovering the capital costs of construction that those costs are in the order of maybe $1 US per cubic meter. >> That, and that's just for the water? >> That is just for water. Now what do we think? The sewer going to be, sewer's going to be higher or lower? Because sewage runs downhill. The smallest diameter pipe in a water supply system is probably two inches in the developing countries. In the United States it's six inches right? The smallest pipe in a pipe sewer system is eight inches. So right away the pipe is much larger diameter. It's much harder to construct. The pipes have to be laid on grade they, so that they're running downhill. So I would expect that the, that the cost of the well run sewage system could be twice as high. It it could be $2 US. >> That includes the waste water treatment. >> That would include the whole ball of wax. >> Right. >> That would be including treatment. So, and the operation of a well run system. One of the problems, one of the reasons why it's difficult to make these generalities is that there is great unevenness in developing countries in maintaining the systems especially when the infrastructure is out of, is underground. It's out of sight you don't see it. So, you know. >> But that's true in industrialized countries too, right? >> It is but, but, but not so much. Even with pipe sewer systems I mean, there are, there are rules now. The reg, when I made reference to the regulatory commissions that watch over these municipal systems. >> Mm-hm. >> That these system have to be, they're, they're using cameras to inspect their sewers. They're reaming out their sewers on an annual basis. They're replacing sewers. That doesn't go on nearly to the same extent in developing countries. So I find it difficult to, to, but I've given you some sense of the numbers. >> Right. Well that's, that's terrific Don. Thank you so much for spending some time with us today. >> It's been my pleasure. >> That's great. [BLANK_AUDIO]
