• Cyburbia is a friendly big tent, where we share our experiences and thoughts about urban planning practice, the built environment, planning adjacent topics, and anything else that comes to mind. No ads, no spam, and it's free. It's easy to join!

Grad / masters 🎓 Better grasp on I-O models, Hedonic regression, multipliers, and travel demand modeling. Is a Masters in Applied Economics the right route to go?

Doberman

Cyburbian
Messages
241
Points
9
I have a somewhat unique background. I've worked in Urban Planning, and Transportation Planning, and now in Economic Development. Frankly, I feel like some of the analysis that I feel like I should be doing is beyond my expertise.
 

WSU MUP Student

Cyburbian
Messages
11,539
Points
52
I have a somewhat unique background. I've worked in Urban Planning, and Transportation Planning, and now in Economic Development. Frankly, I feel like some of the analysis that I feel like I should be doing is beyond my expertise.

I have a masters in planning with a concentration in economic development. The extra economics classes I took towards my concentration were about a 50/50 mix of econometrics and some more in-depth microeconomics fields (particularly around the labor market). I work in the "Economic Development and Community Affairs" department of a large suburban county, and thanks to my education and experience, get to work on a wide range of projects in each of the workgroups of our department (planning/community development, economic development, workforce issues, etc.).

Is a masters in applied economics the way to go?

Sometimes I wish I had gotten a masters or PhD in economics instead of planning, but I'm a nerd and I really enjoy this type of work. That said, IME, that would be overkill at most levels of municipal planning. If you work for a large enough city or county that will care about that type of analysis they will often contract or partner with outside agencies in order to have that third party objective POV (how "objective" that is is a different topic) so you might not use all of your education, especially early on in your career at the lower levels of the ladder.

However, if you wanted to be somewhere like the state labor market/workforce office or at a large COG, that applied economics education combined with experience in planning and economic development would likely be very marketable. Same thing goes for if you wanted to work for a private consulting firm or a research group at a local college or university.

TLDR: What are your end goals, career-wise?
 

Doberman

Cyburbian
Messages
241
Points
9
I have a masters in planning with a concentration in economic development. The extra economics classes I took towards my concentration were about a 50/50 mix of econometrics and some more in-depth microeconomics fields (particularly around the labor market). I work in the "Economic Development and Community Affairs" department of a large suburban county, and thanks to my education and experience, get to work on a wide range of projects in each of the workgroups of our department (planning/community development, economic development, workforce issues, etc.).

Is a masters in applied economics the way to go?

Sometimes I wish I had gotten a masters or PhD in economics instead of planning, but I'm a nerd and I really enjoy this type of work. That said, IME, that would be overkill at most levels of municipal planning. If you work for a large enough city or county that will care about that type of analysis they will often contract or partner with outside agencies in order to have that third party objective POV (how "objective" that is is a different topic) so you might not use all of your education, especially early on in your career at the lower levels of the ladder.

However, if you wanted to be somewhere like the state labor market/workforce office or at a large COG, that applied economics education combined with experience in planning and economic development would likely be very marketable. Same thing goes for if you wanted to work for a private consulting firm or a research group at a local college or university.

TLDR: What are your end goals, career-wise?
Thanks for your response, very informative.

Frankly, I'm not sure. If, for whatever reason, my career sort of stalls in the public sector, I would like to get into the private sector at some point. I have considered consulting or real estate development. I am not under any illusion that I could be some kind of math-wizard at this stuff, but I think being able to analyze say the economic and fiscal impact of transportation infrastructure in development or conduct a market analysis, is a fairly reasonable skill set and goal-based off my background.

I do agree with you that in my limited experience of planning (trans and urban), and econ. dev., I do not see this type of analysis being performed in-house.
 

WSU MUP Student

Cyburbian
Messages
11,539
Points
52
Thanks for your response, very informative.

Frankly, I'm not sure. If, for whatever reason, my career sort of stalls in the public sector, I would like to get into the private sector at some point. I have considered consulting or real estate development. I am not under any illusion that I could be some kind of math-wizard at this stuff, but I think being able to analyze say the economic and fiscal impact of transportation infrastructure in development or conduct a market analysis, is a fairly reasonable skill set and goal-based off my background.

I do agree with you that in my limited experience of planning (trans and urban), and econ. dev., I do not see this type of analysis being performed in-house.

There's also the option of trying to get more experience with existing modeling software and reporting tools. There are tools like IMPLAN that you can use for economic IO modeling that don't require a masters in economics to use, but they do require some training and expertise. EMSI offers a bit more of an elementary modeling experience (based primarily off of the BEA RIMS II multipliers) and they're also starting to get into fiscal impact modeling. Understanding these tools and being able to present your results in way that council/board members can quickly grasp can be a marketable skill as well.

The Federal Reserve Bank of Chicago used to also have an Excel plug-in called FedFIT that did a very sophisticated fiscal and economic impact based on local Census data, local housing costs, RIMS II, commuting patterns, and local property tax rates and apportionments. Unfortunately, they stopped updating the tool a while back but we still use a version I've been updating for our localities.

At the municipal level, sometimes just having a more-than-basic understanding of the concepts behind IO modeling, multipliers, travel demand, etc. is enough. The real valuable part is being able to present it cleanly and succinctly in a way that isn't going to lose the attention of your board members or citizens after a few minutes.
 

Doberman

Cyburbian
Messages
241
Points
9
There's also the option of trying to get more experience with existing modeling software and reporting tools. There are tools like IMPLAN that you can use for economic IO modeling that don't require a masters in economics to use, but they do require some training and expertise. EMSI offers a bit more of an elementary modeling experience (based primarily off of the BEA RIMS II multipliers) and they're also starting to get into fiscal impact modeling. Understanding these tools and being able to present your results in way that council/board members can quickly grasp can be a marketable skill as well.

The Federal Reserve Bank of Chicago used to also have an Excel plug-in called FedFIT that did a very sophisticated fiscal and economic impact based on local Census data, local housing costs, RIMS II, commuting patterns, and local property tax rates and apportionments. Unfortunately, they stopped updating the tool a while back but we still use a version I've been updating for our localities.

At the municipal level, sometimes just having a more-than-basic understanding of the concepts behind IO modeling, multipliers, travel demand, etc. is enough. The real valuable part is being able to present it cleanly and succinctly in a way that isn't going to lose the attention of your board members or citizens after a few minutes.
From what I've gather IMPLAN and RIMS II seem to be the more common modeling program, based on affordability and ease of use. By affordability, it sounds like you have to pay per region with each program. REMI seems very sophisticated and likely beyond what you need financially in a city, and seems to be catered more toward policy, academia, and perhaps the MPO level. Is that an accurate assessment?
 

WSU MUP Student

Cyburbian
Messages
11,539
Points
52
From what I've gather IMPLAN and RIMS II seem to be the more common modeling program, based on affordability and ease of use. By affordability, it sounds like you have to pay per region with each program. REMI seems very sophisticated and likely beyond what you need financially in a city, and seems to be catered more toward policy, academia, and perhaps the MPO level. Is that an accurate assessment?

I would say this is spot on.

RIMS II is very inexpensive but the least robust of the bunch - it's basically just a few hundred dollars for a list of NAICS and multipliers for your region.

IMPLAN is considerably more expensive and would be overkill for most cities but we make it work as we're a county so we use it for projects in 60+ different cities, villages, and townships and also work with our partners to use it for projects throughout the region.

REMI is probably the best of the bunch but we do not subscribe to it just because of the level of expertise that it requires and the expense. However, we do have access to it through our academic partner who has been doing our annual economic outlook for the past 30+ years. REMI can also be used in conjunction with programs like SIMPLAN that can be used to model and forecast population growth, household growth, etc. Our COG that covers a 7-county 4.8 million person region partners with the same university we do to produce the very in-depth long-term regional forecast and its updates every few years.

FWIW, each time our IMPLAN subscription has been up for renewal the past couple of years I've been telling my bosses that we don't need to renew and we could just save the $6k we spend and use it elsewhere. Even with using it for projects throughout the region, I don't feel we get our money's worth out of that one. If you're in a region that is growing faster or really booming, you'd probably get much more use out of these more expensive tools (our development has really slowed over the past few years, especially in terms of big, marquee projects).
 

Doberman

Cyburbian
Messages
241
Points
9
I would say this is spot on.

RIMS II is very inexpensive but the least robust of the bunch - it's basically just a few hundred dollars for a list of NAICS and multipliers for your region.

IMPLAN is considerably more expensive and would be overkill for most cities but we make it work as we're a county so we use it for projects in 60+ different cities, villages, and townships and also work with our partners to use it for projects throughout the region.

REMI is probably the best of the bunch but we do not subscribe to it just because of the level of expertise that it requires and the expense. However, we do have access to it through our academic partner who has been doing our annual economic outlook for the past 30+ years. REMI can also be used in conjunction with programs like SIMPLAN that can be used to model and forecast population growth, household growth, etc. Our COG that covers a 7-county 4.8 million person region partners with the same university we do to produce the very in-depth long-term regional forecast and its updates every few years.

FWIW, each time our IMPLAN subscription has been up for renewal the past couple of years I've been telling my bosses that we don't need to renew and we could just save the $6k we spend and use it elsewhere. Even with using it for projects throughout the region, I don't feel we get our money's worth out of that one. If you're in a region that is growing faster or really booming, you'd probably get much more use out of these more expensive tools (our development has really slowed over the past few years, especially in terms of big, marquee projects).
My understanding is that REMI essentially integrates with the regional travel demand model, is this the case?

IMPLAN seems appealing because I believe it also does a fiscal impact assessment, is that true? So IMPLAN is around 6,000 dollars a year? Is that for one user?

Where and how did you get trained up on these suites? It seems like IMPLAN offers a training program certification, but I don't see anything for RIMS from the BEA.

I appreciate your help.
 

WSU MUP Student

Cyburbian
Messages
11,539
Points
52
My understanding is that REMI essentially integrates with the regional travel demand model, is this the case?

IMPLAN seems appealing because I believe it also does a fiscal impact assessment, is that true? So IMPLAN is around 6,000 dollars a year? Is that for one user?

Where and how did you get trained up on these suites? It seems like IMPLAN offers a training program certification, but I don't see anything for RIMS from the BEA.

I appreciate your help.

I believe you are correct in regards to REMI.

IMPLAN is $6k/year for one user for our regional dataset. I believe the price goes up or down depending on the size (population or number of payroll establishments) in the region you are purchasing the data for. Though I am our user here, I cannot recall all the vagaries of the contract as after I initially hunt down the software I want us to purchase, I send the salesman to our purchasing department and our corporation counsel to negotiate the actual contract and user agreement.

I have not done the official IMPLAN training program and was actually planning to go down to Charlotte for it last spring but then everything shut down. I could do it virtually, but had been putting that off. I've been thinking of doing the virtual version this spring or summer since our office keeps renewing it anyway but I'm trying to hold off to see if my employer starts authorizing travel again as I'd much rather do it in person.

I had a crash-course training session in IMPLAN over the course of a couple mornings at a conference I went to a few years back where they were a sponsor of but I'm sure I'd benefit from their full training program. We also worked with a contractor on a strategic plan who was very skilled with IMPLAN shortly after we first got our subscription and he also helped me out quite a bit and has been willing to answer occasional questions over the past 18 months or so even though our contract for their services has ended.

I don't know of any formal training for RIMS II from the BLS other than the documentation that comes with the multipliers. But the BLS is also pretty good at answer questions via email and telephone (I've found that the staff at places like BLS, Census, and BEA who work on these very specialized programs, tools, and surveys are huge data nerds who generally actually enjoy talking about the products and data that they create).
 

Doberman

Cyburbian
Messages
241
Points
9
I believe you are correct in regards to REMI.

IMPLAN is $6k/year for one user for our regional dataset. I believe the price goes up or down depending on the size (population or number of payroll establishments) in the region you are purchasing the data for. Though I am our user here, I cannot recall all the vagaries of the contract as after I initially hunt down the software I want us to purchase, I send the salesman to our purchasing department and our corporation counsel to negotiate the actual contract and user agreement.

I have not done the official IMPLAN training program and was actually planning to go down to Charlotte for it last spring but then everything shut down. I could do it virtually, but had been putting that off. I've been thinking of doing the virtual version this spring or summer since our office keeps renewing it anyway but I'm trying to hold off to see if my employer starts authorizing travel again as I'd much rather do it in person.

I had a crash-course training session in IMPLAN over the course of a couple mornings at a conference I went to a few years back where they were a sponsor of but I'm sure I'd benefit from their full training program. We also worked with a contractor on a strategic plan who was very skilled with IMPLAN shortly after we first got our subscription and he also helped me out quite a bit and has been willing to answer occasional questions over the past 18 months or so even though our contract for their services has ended.

I don't know of any formal training for RIMS II from the BLS other than the documentation that comes with the multipliers. But the BLS is also pretty good at answer questions via email and telephone (I've found that the staff at places like BLS, Census, and BEA who work on these very specialized programs, tools, and surveys are huge data nerds who generally actually enjoy talking about the products and data that they create).
Thanks.

I may look at IMPLAN. If you're at the county level, I assume; but do not know, that they may be charging you a higher rate based on the larger population and employment counts you have for the entire level. Maybe at the individual city level, the price will be lower.

In your daily activities, what triggers you to run a proposed development through IMPLAN?

Also, at the risk of sounding like an idiot, what is the math behind the model? How do they come up with these coefficients?
 

WSU MUP Student

Cyburbian
Messages
11,539
Points
52
Thanks.

I may look at IMPLAN. If you're at the county level, I assume; but do not know, that they may be charging you a higher rate based on the larger population and employment counts you have for the entire level. Maybe at the individual city level, the price will be lower.

In your daily activities, what triggers you to run a proposed development through IMPLAN?

Also, at the risk of sounding like an idiot, what is the math behind the model? How do they come up with these coefficients?

That is correct on the pricing price goes up or down based on the size of the region (I believe it's population and establishments, not employment, but I might be wrong). Personally, I think you sacrifice a lot in terms of accuracy the smaller the geography that you are looking at when you are running a fiscal or economic impact model - as you clip your geographic boundaries smaller and smaller, you cut off a lot of the laborshed and population that should otherwise be considered in an analysis.

As for when we use IMPLAN - very seldomly do we use it anymore for individual development projects. I'm much more likely to just plug something into EMSI's I/O model since that provides more than enough data for what most boards and commissions can understand if they have a project on the agenda to consider. These days I primarily use IMPLAN to look at the economic and fiscal impacts of particular industries when doing periodic updates of our strategic plans and working on cluster analysis.

Regarding the math: IMPLAN provides a little bit of insight into their models and methodology in their technical documentation but not enough to really figure out what's going on, and what they do provide, they provide in narrative format so you see very little actual math. They keep most of that hidden for proprietary reasons. I'm pretty sure that if I contacted our account rep they'd put me through to one of their technical people to give me more (but still limited) background. However, I sort of like not digging into it too deeply as a) I'm no longer interested in sitting down to try and build my own models and b) being able to basically say to our boards and commissions that, "These forecasts come from the leader in I/O modeling and our contract does not provide us access to their full methodology" gives us a nice buffer. As a county in a state where counties don't actually approve development projects this has worked fine for us. Even when performing analysis for our individual cities, I think I've had their staff contact me for more information maybe 2 or 3 times in a decade, and even on those occasions, their usually looking for more in-depth data or a "what if" scenario changing some of the assumptions.

As with any modeling or forecasting, I think it's so important to make sure that your audience understands that the forecasts are just that - a forecast and they rely heavily on the assumptions (inputs) that are put into them. So if tax rates change, commuting patterns change, some other big development happens (or fails), there's a natural disaster, etc. the actual real world results will change. That's why at the end of any good forecast presentation there is usually a few minutes spent on risks to the forecast. CYA.
 

Doberman

Cyburbian
Messages
241
Points
9
That is correct on the pricing price goes up or down based on the size of the region (I believe it's population and establishments, not employment, but I might be wrong). Personally, I think you sacrifice a lot in terms of accuracy the smaller the geography that you are looking at when you are running a fiscal or economic impact model - as you clip your geographic boundaries smaller and smaller, you cut off a lot of the laborshed and population that should otherwise be considered in an analysis.

As for when we use IMPLAN - very seldomly do we use it anymore for individual development projects. I'm much more likely to just plug something into EMSI's I/O model since that provides more than enough data for what most boards and commissions can understand if they have a project on the agenda to consider. These days I primarily use IMPLAN to look at the economic and fiscal impacts of particular industries when doing periodic updates of our strategic plans and working on cluster analysis.

Regarding the math: IMPLAN provides a little bit of insight into their models and methodology in their technical documentation but not enough to really figure out what's going on, and what they do provide, they provide in narrative format so you see very little actual math. They keep most of that hidden for proprietary reasons. I'm pretty sure that if I contacted our account rep they'd put me through to one of their technical people to give me more (but still limited) background. However, I sort of like not digging into it too deeply as a) I'm no longer interested in sitting down to try and build my own models and b) being able to basically say to our boards and commissions that, "These forecasts come from the leader in I/O modeling and our contract does not provide us access to their full methodology" gives us a nice buffer. As a county in a state where counties don't actually approve development projects this has worked fine for us. Even when performing analysis for our individual cities, I think I've had their staff contact me for more information maybe 2 or 3 times in a decade, and even on those occasions, their usually looking for more in-depth data or a "what if" scenario changing some of the assumptions.

As with any modeling or forecasting, I think it's so important to make sure that your audience understands that the forecasts are just that - a forecast and they rely heavily on the assumptions (inputs) that are put into them. So if tax rates change, commuting patterns change, some other big development happens (or fails), there's a natural disaster, etc. the actual real world results will change. That's why at the end of any good forecast presentation there is usually a few minutes spent on risks to the forecast. CYA.
This might be a back-end question, but are these models integrated with GIS at all? I believe REMI or at least the travel demand model. For example, how are these multipliers taken into account from one city to the next or into the county as you alluded to? Or are they just sort of in a spreadsheet or Access, linking one adjacent city to the other depending on the size of the development.

Would a rezoning case trigger an impact analysis? For instance, going from AG to Manufacturing as part of the justification?
 

WSU MUP Student

Cyburbian
Messages
11,539
Points
52
Not sure about integration with GIS but I assume IMPLAN and REMI both have those capabilities to some extent. In essence, geography matters and they are using a lot of Census and BLS data in these models so I imagine there are FIPS codes associated with those geographies on the backend.



As a county, we rarely ever get involved with rezoning (not a thing counties in Michigan generally handle) so I haven't done any analysis specifically for rezoning cases. But that said, when I do run a model for a potential new development, I'm usually told some things like Company X wants to locate at Geography Y and bring Industry and Employment Z. City council in Geography Y wants to know what sort of fiscal and economic impacts might be associated with it. I will look up the specific location and get property tax rates and apportionments and usually I just assume that the correct zoning is in place. However, thinking back, it's possible that these development proposals are part of the justification for (or against) a rezoning. In fact, it's more than likely probable that that's been the case many times.
 

Doberman

Cyburbian
Messages
241
Points
9
Not sure about integration with GIS but I assume IMPLAN and REMI both have those capabilities to some extent. In essence, geography matters and they are using a lot of Census and BLS data in these models so I imagine there are FIPS codes associated with those geographies on the backend.



As a county, we rarely ever get involved with rezoning (not a thing counties in Michigan generally handle) so I haven't done any analysis specifically for rezoning cases. But that said, when I do run a model for a potential new development, I'm usually told some things like Company X wants to locate at Geography Y and bring Industry and Employment Z. City council in Geography Y wants to know what sort of fiscal and economic impacts might be associated with it. I will look up the specific location and get property tax rates and apportionments and usually I just assume that the correct zoning is in place. However, thinking back, it's possible that these development proposals are part of the justification for (or against) a rezoning. In fact, it's more than likely probable that that's been the case many times.

I had a demo with IMPLAN today and was pretty impressed. My layman's take is that RIMS is a little too basic, REMI is too academic, and IMPLAN is a middle-solution.

They gave me a quote of 2500 for one county, 3500 for two with associated zip codes included (I'm assuming if the zip codes overlap county boundaries they are included entirely). We are partially located in two counties, particularly some of our major commercial areas so I was planning on pitching two counties at minimum.

If we are a medium sized suburb, with most of our gainful employment going into the city, should I look at a subscription that covers the city as well? The city is not located in either of the counties that we are located in. Should I look at getting our entire metro area?
 

WSU MUP Student

Cyburbian
Messages
11,539
Points
52
I had a demo with IMPLAN today and was pretty impressed. My layman's take is that RIMS is a little too basic, REMI is too academic, and IMPLAN is a middle-solution.

They gave me a quote of 2500 for one county, 3500 for two with associated zip codes included (I'm assuming if the zip codes overlap county boundaries they are included entirely). We are partially located in two counties, particularly some of our major commercial areas so I was planning on pitching two counties at minimum.

If we are a medium sized suburb, with most of our gainful employment going into the city, should I look at a subscription that covers the city as well? The city is not located in either of the counties that we are located in. Should I look at getting our entire metro area?

I'm not sure the size of your metro area, but for stuff like this, I always recommend, at a minimum, getting the data for your entire metro area as that should cover your entire laborshed which I feel is especially important in any analysis that also takes commuting patterns into consideration.

If cost is a concern, IMPLAN may allow you to partner with another organization and buy two licenses at a slightly reduced cost per license (for instance, instead of one license @$3,500 for the two counties of data, you buy two licenses @$5,000 and you pay $2,500 and maybe somebody from that other county or from the local Workforce Investment Board or community college pays the other $2,500).
 
Top