Thursday, September 29, 2011

Real Estate Market Analysis

Current Economic Conditions
America is in the third year of a ‘Great Recession’ brought on by a financial crisis. Depressions and recessions induced by financial crises tend to be longer lasting than other recessions. It takes time for banks to de-leverage, write of bad debts, and attract enough deposits to restore their core capital ratios to sufficient levels to begin making additional loans, a process largely dependent on the population savings rate.  Until that occurs, the supply of private credit to business is limited. Given the current popularity and financial advantages to debt financing, ongoing sources of credit are no longer essential just to business expansion, but also integral to business operations and cash flow management.  After the collapse of Lehman Brothers, the climate of uncertainty slowed both the expansion of the money supply and the velocity of money, resulting in a ‘credit shock’ that rippled through out the economy, resulting in delayed payments and missed payments. While the Federal Reserve policy of QE (Quantitative Easing) has increased the volume of money available to overcome the lack of liquidity, political opposition to it is increasing, and it seems unlikely to continue.
The recession has also resulted in massive job losses and an increasing unemployment rate. While job creation is increasing, it is currently only keeping pace with the rate of population growth, and is thus insufficient to bring down the unemployment rate. Additionally, because unemployment only counts those seeking a job, official statistics fail to recognize the ‘shadow pool’ of previously employed workers.
Local economic conditions are rosier. The State of Utah has endured the recession far better than many states, with ongoing economic growth and an unemployment rate lower than the national average. Despite also having been victim of the ‘housing bubble’, its rate of house-hold creation exceeds the national average as a result of its higher population growth, and will be able to reduce the stock of unsold housing at a faster rate.  

18 Month Outlook
The eighteen-month outlook continues to look grim. Previous economic recoveries were ‘housing led’, an impossibility under current circumstances. As a result of both a growing population and a declining household size, new households were continuously being formed. New households represented an ongoing demand for additional housing units. That demand in turn fueled residential construction, and related household formation induced purchasing (furniture, home construction, appliances, etc.). The construction industry in turn provided employment with low skill, low income laborers with low education. Further, this led to a competitive shortage of labor for other low-skill positions, resulting in rising wages. Given the propensity of low-income workers to spend a high proportion of their income as consumption, this resulted in a ‘virtuous cycle’ of economic activity. That dynamic will not be pushing us out of this recession.
Previously able to resell mortgages to be repackaged into mortgage backed securities, banks are now stuck with any mortgages they originate, a trend pending regulation seems determined to reinforce. As a result, lending standards for mortgages are tightening. Combined with the need to rebuild core capital (as mentioned earlier), and declining income, the pool of qualified mortgage borrowers has shrunk substantially.
Further, the rapid appreciation in housing prices permitted many Americans to take home equity loans. These loans were then used to fuel consumer purchases, artificially elevating retail sales per square foot, and raising rents. Responding to these artificially high rents, additional retail space was constructed. Following the advent of the recession, sales tax records show consumer demand crashing, reducing the demand for retail space. Demand on retail space cannot be expected to return to the pre-recession ‘normal’.      
Again, the outlook for Utah remains rosier. Because of its larger than average family size, and high dependency ratio, a higher percentage of purchases are non-discretionary. While inferior complements (cheaper toys, off-brand food) can by substituted, they actual purchase cannot be avoided.
Utah is also riding a rising tide in the health care and education sectors. While the future seems likely to hold lower federal payments for medical care, the aging population of baby boomers will continue to demand (and pay for) quality medical care. Another consequence of the recession has been the return people (especially young adults) to school, and to keep them there longer as they wait out a bad economy. While this has future implications on  National financial stability, the present bounty in education is fueled by subsidized Federal loans, a dynamic which seems unlikely to change.

Real Estate Market Analysis Methods
The purpose of real estate market analysis is to reduce risk in real estate investment. The primary threat to profitability is competing supply. When supply exceeds demand, prices fall to compensate, and may result in a negative capitalization rate. To ensure that real estate development is profitable, it is necessary not only to estimate existing supply and demand, but to be able to forecast future supply and demand at the end of the real estate development process, so that new product can be absorbed into the market at an acceptable rate.
While there exists a large stock of real estate, demand for new real estate is driven by the small segment of businesses and households currently seeking new or additional space. Unlike many industrial products, one of the essential elements of real estate development is developable land. And, as the saying goes ‘They aren’t making any more of it’. While many parcels are interchangeable, parcel locations are (if not unique) then certainly limited. As a result, real estate demand takes place not at an aggregate level, but at a local level.
Critical to determining both demand and supply is the market area for which both totals are drawn. The market area varies by product type. The market area for a regional resort is different for that of a convenience store—the former may span several states, while the latter may only span a few blocks.
There is never exact comparability between product types—in order to avoid niche saturation and direct competition many developers attempt to bridge niches, establish cross-niche appeal, or provide a superior package of amenities. As a result, even with the same geographic area, two different analysts may produce different supply and demand numbers.
Additional supply of a product type are created primarily by competing developers, but may also originate as a result of conversion or rehabilitation of existing real estate. As mentioned earlier, not only the current stock must be assessed, but the ‘pipeline’ representing expected future supply of currently planned, permitted, or under construction.
            Demand for real estate is a more complex phenomenon, because it tends to be more elastic. Increasing the usage of existing space (through over-crowding or more intensive use) can be increasingly inefficient, but it faces no ‘hard’ limit. For each major product type (residential, retail, office, industrial, and service), there are different drivers of demand. For residential, it is a combination of household formation, financing, and socially conditioned tenure-style expectations. Demand for retail and service real estate results when the expected rents per square foot exceeds the construction cost per square foot of additional space. Demand for office space tends to be conditional on increase in employees in the types of businesses that demand for office space, with different types of office users demanding different ‘classes’ of office space. Industrial is similarly driven by sector growth, but is also contingent upon access to necessary inputs—proximity to raw materials, processing facilities, supplier clusters, or equivalent transportation access. Data for projecting employment growth by sector can usually be obtained at a state or metropolitan level from the relevant business/demographic/planning agency or body.

Wednesday, September 28, 2011

The Future Holds..

Utility access (especially transportation access) determines the development potential of many parcels. Historically, a steady supply of developable land has been ensured as a result of an expanding roadway network. Federal subsidy for road network expansion can be expected to become scarcer and Federal funding is reoriented towards preserving existing roadways. Freeway and highways, which provide the greatest increase in ex-urban accessibility to urban employment and services, have typically been justified on the basis of congestion mitigation. Future efforts on congestion mitigation are more likely to emphasize less costly and more efficient solutions such as ITS and transit improvements. While roadway expansion can be expected to continue, it seems likely future roadway expansion will no longer exceed the rate of population growth. 

Friday, September 23, 2011

American Community Survey

Replacement of the Census long-from with the American Community Survey has destroyed the capacity to perform multi-census analysis of many data series. Many factors for which it was previously possible to make highly reliable estimates even for small geographies, it is no longer possible to do so. The future holds great promise for those that can combine the data collected in the 2010 census with the data available through the American community survey to make accurate inferences about the broader population.

Wednesday, September 14, 2011

HSR not a panacea

An excellent article Martin Buchert turned me on to:

To some extent, HSR is a 'solution looking for a problem'. Like the advantage of LRT over bus, some of the speed gains are created purely by bypassing stops* and concentrating activity into highly networked hubs.

Which makes me wonder about the new UTA bus system. UTA has unveiled a series of 'flex' buses that are demand-response (dial-a-ride**) rather than fixed route buses. It seems to be positively received, and will (hopefully) lead to fewer empty buses meandering through suburban neighborhoods in an effort to fulfill federally mandated coverage standards.

Perhaps the long-term evolution of transit network will be a 2-tier system--Flex shuttles ferrying people in and out, and long-straight high speed BRT and TRAX lines moving people from location to location. The key issue becomes 'peak hour' transportation, when everyone is trying to use the system at once. Make the 'Flex' shuttles larger, unto 40' buses? Or run two shuttles?

*God knows Amtrak could use some of that.
**How is that not a private taxi? -- Because it only drops you within its service area, or at the nearest transfer point.

Tuesday, September 13, 2011

Photo Morphing

I attended a technical presentation by Fregonese Associates at Envision Utah on doing photomorphs--aka changing streetscapes over time.

Photomorphing--it doesn't have to be that good--just good enough to fill the eye, and let the brain fill in the rest. Start with a good image--maintain consistent lighting--minimal shadows. Picture big enough to convey changes in roadway right of way and streetscape. Enough 'blighted' or underused private property that can reward changes over time. Can shoot down centerline of street, or from one of the sidewalk corner, at edge of property lane. A 30-60 angle, to see down one street, to explore nature of pedestrian environment.

Add buildings--take a picture of a building you like, or cut it from your library--get perspective right, to get vanishing points right, take picture from the same angle. Use the transform features in photoshop to distort the building so it matches the picture.

Same for people and cars--rough cut out, including shadow, scale them basically right, pop them in. To scale people, use the height of the building. Follow vanishing lines. Taller buildings a little more complicated, because of lens distortion, but not much.

Spend the time to get shadows in, even if poorly done. To make a shadow--take a copy of the image, adjust brightness to zero to make them gray, then reduce transparency, then use the transform tool to distort it and squish it down so it sits and the feet.

Frego has entire LIBRARY of people and buildings and objects. Frego uses their pictures and cutting out the people. .png format. Frego has a library of about 30 people that they just use and re-use. Commercial Version is photoshop plugin that provides a library of trees and people and images, including rotating people to many angles.

Cars can be tricky--may need to remove some cars. Use clone tool to clone nearby area, and paint out the other area. Follow the same vanishing plane. Aligned--pull from same distance from the point. Aligned--duplicate from original alt-selected point. Try to keep color profile identical to surrounding area.

Clean up the base image of temporary or obstructing items (cars, cones, power lines, streetlights). Use shadows/highlights to brighten things up--balances highs and lows of over and underexposure. Can automate a batch function in photoshop to do a series of tasks.--like records a macro in excel.

Corner bump-outs a big deal--reducing walk-out distances to cross the street. Wider streets actually SLOW traffic, because the width of rode requires longer pedestrian signals, due to ADA accessibility.  30 seconds on Seventh East.

Dynamic Travel Assignment

WFRC is building a new traffic model. I'm attending the update meeting, and taking notes. Thought I might as well publish them. I believe they are using Dynus-T.

Ability to model how people shift their routes or departure times in response to congestion, pricing, controls, incidents, and improvements. Also, understand how individual impacts affect entire region. Not as detailed as micro models, but capable of high-fidelity simulation of entire region. Not micro-simulation of individual cars.

Macro-Meso-Micro integration.
Travel Demand Model-->VISUM-->VISSIM
 Automatically integrating data from the higher tier into the lower level of modeling as part of the model flow.

NEXTA.exe. Fort Worth Network in Dallas, 13 zones. Each link with data loaded.
Dynamic traffic assignment--can see traffic density over time.

--Transit network built on highway network. Cube model uses that info to extract data about how long transit trips will take, and how many people will take transit on that basis. But the percent of people who have walk-access to the transit station is a big question--what portion of pop/emp in a TAZ can do this?

WALKPCT identifies the percent of a zone that can actually walk to transit.

Local Bus--buffer around a line
TRAX/BRT--Buffer around stops.