ASTM just published changes to their E2026 and E2557 standards for Seismic Risk Assessments and PML reports. Chief among the changes are more stringent requirements for who can conduct and review certain seismic risk assessments. This is just the latest in a string of recent shake-ups in the seismic world, including mandatory retrofit ordinances that were implemented in Los Angeles and San Francisco, changes to agency seismic policies, and the inclusion of man-made earthquakes in the USGS hazard map.
How will these changes impact lenders and investors? Should you revisit your seismic risk policy? Learn more in this webinar hosted by Partner and GlobeSt.com.
Panelists Joe Derhake, PE, Bill Tryon, and Jay Kumar, PE will address:
Jeff: Hi, I’m Jeff with globest.com. Thank you for joining us for today’s webinar about managing seismic risk in today’s regulatory environment which our speakers will address recent ASTM updates, retrofit ordinances and other seismic policy considerations. A few housekeeping items before we get started. At the conclusion of the webinar, you will be able to download the presentation slides which will include links to additional resources. You can submit questions via this console which will be answered by our presenters following the webinar. We will also provide the speakers’ contact details at the conclusion of this webinar so that you can reach out to them directly.
This is a joint effort between globest.com and Partner Engineering and Science, who are one of our thought leaders. This webinar is of course owed to the efforts of our three expert presenters. First off, we have Joe Derhake, CEO of Partner. Joe hosted a very successful webinar about seismic risk assessments through globest.com several years ago and will today provide context for changes in the seismic landscape and how this should be addressed in lender and investor risk management policies. Jay Kumar is Partner’s Technical Director for Structural Engineering and Bill Tryon who served as ASTM chair for the E2018 standard is Partner’s Director of Strategic Development. With that, I will hand it over to Bill to introduce today’s agenda.
Bill: Thanks. First of all, thanks everyone for joining us today. Partner as a company has been trying to promote clarity around seismic risk evaluation since 2010. And there have been some recent changes to ASTM standards that we think make this a good time to revisit the topic. We’ll start with some basics about estimating seismic risk damage, some recent updates to ASTM standards for PMLs, mandatory retrofit ordinances that create additional risks and then we’ll talk a little bit about policies and the things that you can consider when you write or update policies.
Some of you may already know me, but for those who don’t, I’m the Director of Strategic Development for Partner, Chair of the ASTM Committee for Property Condition Assessments, and have participated in the creation and revision of several other standards in ASTM including the two PML standards we were going to talk about later today. I came into consulting in 2007 but before that, I was in charge of Physical Due Diligence for Wells Fargo for about 20 years. Wells is based in San Francisco and has been evaluating its seismic risks longer than most lenders.
After the Loma Prieta quake in 1989, Warren Buffett, who was Principal Investor at Wells Fargo, asked our CEO about earthquake risks. Since I was running our technical group at the time, the question filtered down to me. Originally estimates of PML were used by insurance companies to set financial reserves and rates for fire insurance. The concept was extended to earthquake losses in the ’80s. When this question came up, we did some research and began using the software developed at Stanford to get a better understanding of our risks. The PMLs for earthquakes weren’t really used much in the real estate due diligence world until the mid-’90s when CMBS became concerned about the effect of earthquake risks on performance of loan portfolios.
Wells was a pioneer in the CMBS model. And for a year or so, we actually performed internal PML calculations for properties and portfolios until the agencies asked us to consider using third-party evaluation. Back in the ’90s, there was a lot of confusion about PMLs. The terms SEL and SUL hadn’t been defined and most people didn’t understand that confidence levels for both ground shaking and building damage could affect the PML estimate. Because CMBS transactions involved portfolios of loans, Wells adopted an approach to PMLs based on the 90% confidence level for ground shaking and the median building damage. Which is what the ASTM eventually recognized as the common approach to CMBS transactions.
So, I’d like to turn it over to Joe Derhake for a second. Joe is the CEO of Partner. He’s an engineer and he does an exceptional job of explaining PMLs and how they work. But before we go into calculations, I’m gonna ask him to talk a little bit about some other changes to the seismic risk landscape. Joe?
Joe: Thanks, Bill. So, you know, our firm does environmental assessments, property condition assessments, to ALTA surveys and seismic risk assessments in support of due diligence. Overall those trades that we participate in, I think there’s the most opportunity, well at least I felt this way maybe five years ago, the most opportunity to improve consistency as a whole industry in seismic risk assessments. Because I think our clients sometimes were frustrated that they get different answers from different engineers.
So, five years ago, we set out, trying to prove this, and then this new ASTM standard that just came out you know, weeks ago is a big step to helping the engineering community to assess buildings more consistently. But definitely half of the formula, in my opinion, the other half of the formula is user responsibilities. And it really helps if we are able to educate our clients not so much on how to be a structural engineer at all, but just to understand the vocabulary and what we’re talking about here. So, you can ask the right questions of the engineering community, ultimately yielding more consistent advice.
There’s been a lot of earthquakes and that they’ve…you know, the Northridge earthquake which is a 6.3 magnitude earthquake, produced $40 billion in damage and killed 57 people. Ithit in the middle of LA, and Northridge is actually an area of the city of LA. You know, that 6.3 isn’t a particularly large earthquake. If it was a 7.3, it’d be 10 times as big in terms of wave amplitude, 30 times as much energy would be released. But the question is how much more dollars of damage would it cause to real estate? And it’s hard to say really, but to say $400 billion, ten times as much. You know, of course, the $40 billion number was 1994 economics and 22 years later, everything’s more expensive. So, I believe that our industry as a whole would be smart to invest in making our buildings a lot more prepared. I mean if the west coast put $10 billion in retrofit work into its building stock, perhaps we could halve the amount of damage we experienced. Before we get into you know, talking about what actually changed in the ASTM standard and retrofits, let’s go into…let me just kind of explain 101 on how we assess seismic risk.
The very basic inputs to a PML are: the strength of the building, that’s kind of obvious; the type of soil the building sits on; And the expected lateral loads that the building will experience, in other words how much shaking the building will experience will determine how much loaded experience is. We usually model buildings in the mortgage industry to the 475-year event. So just to get your head around that real quick, everybody I think understands the concept of a 100-year flood, right? Some, you know, high-water mark literally in the next hundred years. Well, if there’s 100-year flood there’s 100-year earthquake, and there’s a 500-year flood and a 500-year earthquake. We modeled with a 475-year earthquake which seems like an odd number to people, but the way the statistics work out, there’s a 10% chance of that occurring in 50 years. You’d think it’d be 500, but the way the statistics work out, it’s 475. You will see a PML occasionally modeled to a different time period which makes it a very non-standard product for the mortgage industry.
But insurance guys would do other things. And so, that’s something to be wary of if you ever see two reports with you know, radically different estimates. When we model a building the engineers really don’t model and predict a single number, it’s almost a curve of numbers. A bunch of scenarios, each have a certain probability of coming true. Like you see here, this is a curve. And so, a lot of different things could happen. It could be very bad, it could have almost no damage. And there’s for sure, a giant randomness to how buildings perform and how much damage they receive. The middle of that curve is something we call the SEL, the Scenario Expected Loss and the SUL is the Scenario Upper Loss which is the 90% confidence level. You would only have a 10% chance of going over that level in the next 475 years. Or if a 475 years’ event came to pass.
You know, a lot of people get caught up on “Do I want my PML to report the SEL or the SUL?” I would answer that question that for the most part if you’re comparing the number to 20%, you really want to use an SEL. There are people that don’t follow that rule of thumb, but for the most part, that is the industry standard. The relationship between the SUL and SEL has also been a real controversial issue through the new ASTM standard 2557 which we’re going to talk about in a moment, offers this curve which really kind of expresses an expected spread between the SEL and SUL. And you notice in this curve, the more work you do, the tighter that spread would be.
If you’re doing a desktop, that spread would kind of be high because there’s more uncertainty. And if you do a more detailed study, then you’d expect you’d know more and therefore it’d be often that the SEL and SUL would be closer together. So, let me walk into how we calculate a PML. First of all, ASTM, the ASTM standard does not provide a method of calculation. In fact, that’s one of the main sources of inconsistency, engineer to engineer. There’s more than one reasonable way to calculate it. And maybe we need an earthquake to see which one’s the best. Here I’ve listed in the bottom several fine ways to calculate a PML. I’m going to walk you through one of them, which I believe is the market leader and used most often in the mortgage industry. And that’s Thiel Zsutty method. It is probably the best method for explaining to laypeople how we calculate the PML.
There’s really three variables for the actions. The first one is ground acceleration. This is one where engineers should agree. You can go in the USGS map and it’ll give you a number, like for example, 0.5 Gs. So, the a-value would be 0.5 in this circumstance and that means half of the force of gravity would be going sideways in that quarter, in 475-year event as predicted by the USGS, United States Geologic Service.
The next variable is the S-value. And you know, there’s people who say, “Hey look, we have two buildings that are almost the same but you gave me two different PMLs.” Well, of course, they’re in different places. They’re sitting on different soil and they’re going to experience different accelerations. This chart I have here is somewhat illustrative of the effect soil has on shaking.
This is the Loma Prieta quake, it happened kinda near Santa Cruz and south of the Bay Area. But let’s say you start going directly north of where the Loma Prieta quake hit, you’d see a lot of yellow and green which represents lower shaking, red and orange is higher shaking. But if you go northwest, right up through the Bay Area, you’ll see a lot more orange and red. And some of the most severe impacts as a result of the Loma Prieta quake came in Oakland. And that was because of the soil type. First off, shallow groundwater amplifies shaking. And secondly, it matters what type of soil you’re on. Here’s a chart from the American Society of Civil Engineers, that talks about different classes of soil and how stable they’d be.
What we would do is when we’re doing a PML is assigned a S-value, based on this information. Sometimes the soil has a specific issue, like for example, liquefaction and in a liquefaction, it could be sandy soil and it would turn into quicksand for just a few seconds which of course is a problem for a big heavy building. Liquefaction zones are well mapped in California and to some extent well mapped in other states. If your property is in a liquefaction zone, it’d be appropriate for the engineer to penalize the PML with a high S-value. In all instances, these zones are really the state’s best estimate of where liquefying soils may exist and a site-specific geotechnical report would be superior to these maps. Sometimes that’s why you’ll see us change a PML if we get a geotechnical report because we’re able to not penalize the site for liquefaction.
The last variable I want to talk about is really where the action is and that’s the building coefficient. Obviously, different buildings will perform, and some are strong, and some are weak. And we will assign a b-value between 0.1 to 1.25. The Thiel Zsutty method offers a big chart with different b-values. Here you see a chart with a bunch of what would be, there’s maybe there’s 48 different categories of Thiel Zsutty. And what they did was they were able to study a building type and ultimately produce a b-value because we have enough data to make a judgment on how those buildings would perform. It’s almost like an appraisal system where you say, “Hey look, this building is comparable to this set of buildings that we study.” And that allows you to assign a b-value. But just like appraisers, if your neighbor’s house sells for $300,000, you might say, “Well gee, my house is about the same size, it’s worth $300,000.” But the appraiser might make an adjustment because maybe you have a pool. And so, the appraiser says, “Well, you have a pool. So, we think you’re worth $330,000.” The engineer would follow a similar process and they’d say, “Well, hey look, these buildings are very similar, if you have a moment resisting steel distributed frame, low rise building we’re going to assign a b-value 0.11. However, it has a big cantilever and that would create some torsion and irregularity and therefore we will assign a little bit higher b-value to penalize the building for that deficiency.”
Here’s a list of, I’m not gonna show you the whole table, but here’s what the bottom of the table looks like and these would be higher risk buildings and you can see that the b-values are much higher and would cause higher PMLs. That’s kind of 101 on how we do this and I think users are well served to be somewhat conversational on the map. I would like to turn it back over now to really what’s prompting this whole webinar and that’s the changes in the two ASTM standards. Bill, you want to take it back over?
Bill: Thanks, Joe. Let’s get a look at the ASTM standards for a minute. ASTM produces two types of standards or several types of standards. Guides and practices are the most common. And the PML standards are one of each, one guide and one practice. They published the first guide for seismic evaluation in 1999. And essentially, the guide provides a framework for evaluating seismic concerns. It discusses key concepts of PMLs, defines terms and identifies levels of investigation but it doesn’t, as Joe said, establish specific requirements for evaluating PMLs or calculating PMLs. By the time the standard was published in 1999, PML really had become a misunderstood and a misused term throughout the industry and this first version of the PML guide actually discouraged using the term PML at all going forward.
In 2007, the guide was revised and ASTM also published the E2557 Practice for PML Assessments. The practice does acknowledge PML as a valid term and it leverages definitions, tasks, and levels of assessment from the 2026 guide but boils those down into specific steps and criteria to estimate PMLs. The standard allows users to decide on their own definition of PML, but notes that the SEL estimate as Joe mentioned earlier, is normally used for CMBS transactions. Both standards were realized earlier this month since ASTM requires renewal of standards every eight years. There are some exclusions to the 2557 practice. It’s concerned primarily with estimating building damage and evaluating the site and building stability of properties.
But there are other risks that can result in losses for investors and lenders. Things like regulatory requirements, damage to contents and inventory, building continuity, beneficial use of the property, fire damage from earthquakes; regional impacts aren’t covered by 2557. They can be addressed, but they typically need to be added to the scope of the PML. This is a really important point. Sometimes clients order a PML without really understanding what it means.
Revision of the standards reemphasize the importance of understanding what you need and ordering the right thing. According to the practice, users should establish acceptable levels of uncertainty, decide how they want to measure damage, consider how budget and time constraints will affect uncertainties and decide what levels of investigation they need. Users are also required to identify the condition of the property, describe any retrofit to be considered by the engineer, evaluate the engineer’s qualifications and determine whether evaluation of additional concerns that we just mentioned is needed.
So, what else has changed? There are a lot of other revisions and standards but we’re only going to talk about the few that we think are most important. For example, some definitions were refined or added to the standard. The definition of user was revised but is substantially the same as the prior version. The user is the party that engages the engineer to perform the work. Definitions were added for two consultant goals. The Field Assessor is a person who visits the property, who collects property information. The Senior Assessor is a person who has responsible charge of the assessment. This becomes really significant in light of another change to the 2026 guide consultant qualifications.
The prior version of 2026 provided guidance on qualifications of consultants but was subject to different interpretations. Current revisions are more definitive. The Senior Assessor must be a licensed civil or structural engineer and have 10 years general structural engineering experience, five years seismic design analysis and three years seismic risk analysis experience. The field assessor must also be a licensed engineer but can have less experience. We think this is the most important change in the recent standard. It doesn’t affect screening levels of assessments, what ASTM calls a level 0. But level 1 and 2 assessments are used by most of our clients.
Historically, some level 1 assessments were based on property observations by consultants who performed environmental or property condition assessments for properties. Under the revised standard, field observations have to be done by the licensed engineer with experience that satisfies the new standards. And that will frequently not be the same person you want to evaluate the condition of your roof or the facade or other improvements of the property. Mostly, consultants will have to send another person to the site and that person will not be cheap. Our structural engineers bill out at nearly $200 an hour. The revisions do allow some exceptions for this. As I mentioned, they don’t apply to level 0 assessments. For level 1 assessments and above, the Senior Assessor can modify qualifications for the Field Assessor when they have structural plans for the building or the building was built after or during a benchmark year, and when there have been no significant structural changes to the building.
Here’s an example of benchmark codes taken from ASCE 41. Construction during or after a benchmark year allows the consultant to assume certain things about construction. Benchmark years are different for different property types. Essentially a benchmark year is the point at which building codes for a particular type of construction are assumed to have corrected the most common seismic design deficiencies as we understand them. Buildings built during or after a benchmark year are assumed to comply with codes and the Assessor may decide that observation by an engineer is not needed. So, this is an automatic.
There was also a change in standard conclusions required under 2557. In the prior version, the consultant was required to include a conclusion statement that the report was in compliance with the 2026 guide and the 2557 practice to identify the level of assessment. Note the section including any deviations from the standards and provided PML conclusions and definitions. The standard though also requires that every report address the site and building stability. These things weren’t addressed in the conclusion statements. And revisions of the standard do now include specific statements regarding building and site stability.
There are lot of other changes in the standards. For example, two new forms are included in the appendix 2557. Form X4 is a summary of key information from the assessment and X5 provides some generic information that should help users understand the limitations of the assessment. The standard says these forms should be included rather than must, so they might not be included verbatim in all reports. Other changes also reemphasize the importance of reviewing plans. You know, so much is inaccessible during site reconnaissance, but plans are by far the best source of information about buildings. Users with a low risk tolerance or hoping to avoid a site visit by an engineer and save a few dollars should be especially careful to make sure the plans are provided for review.
Another change is the relative to third-party technical review. Third party review is encouraged throughout the standards, and 2026 includes a definition of third party review and discussion of the qualifications of the reviewer. And 2557 requires that reports be sufficiently detailed to allow the independent review to evaluate the work. The prior version of 2557 included guidance on levels of investigation. But a new appendix has been added that can give users a better understanding of earthquake risks and management of uncertainties. Appendix X1.2 includes some great information on site failure, damage measures, and uncertainties and also reemphasizes the importance of reviewing plans in geotechnical reports as well as selecting the right consultant to do the work.
There are also some things that didn’t change that we think are worth talking about. Appendix 2.2 of 2557 includes this map. It was originally published as part of the Uniform Building Code in the 1990s. Investors and rating agencies began to require lenders to perform seismic risk assessments for buildings in Zones 3 and 4 shown on this map. Here’s a color-coded version of the map that you may be more familiar with. Map was dropped from the codes more than 15 years ago but it’s included in the appendix of 2557 as a screening tool. This map is not included in the standard. It’s published by USGS and shows the PGA or Peak Ground Acceleration based on 10% chance of exceedance in 50 years which is the same probability used for most PMLs.
Some lenders like Fannie Mae and Freddie Mac have started to use the PGA as a threshold for acquiring PMLs instead of the seismic zone map. This version of the map highlights the major differences between the zone map and the PGA map, so it gives you a good idea of the differences. At the end of the day, the seismic zone map is based on the state of science 20 years ago. But information for peak ground acceleration is more current and is easily available from USGS. So, PGAs provide a more accurate screening tool. Some industry requirements though like the CREF C reps and warranties refer to the seismic zone map. So, some clients may do better to follow the zone map rather than switch to PGA maps.
I also want to touch on increases in earthquake activity in some areas. The number of earthquakes has been increasing in the areas highlighted by red rectangles on this map. Earthquakes in these areas apparently result from underground injection of wastewater resulting from oil and gas production. This map only includes peak ground acceleration from naturally occurring earthquakes. This map is from a USGS study published earlier this year and shows estimated ground shaking for both human-induced and naturally occurring earthquakes in the next year. You can see that Oklahoma is really now the capital of earthquake activity in the continental United States though they haven’t had any earthquakes with widespread damage at this point.
These two maps give a little better understanding of how dramatic the increase in earthquakes has been. The map on the left shows the number and size of earthquakes in Oklahoma in the entire decade of the 1980s. The map on the right shows earthquakes so far, this year. Evaluating PMLs in these areas though is tricky because the frequency and size of earthquakes can’t be predicted based on the geologic record. And building codes haven’t historically incorporated seismic design requirements equivalent to California buildings. Even in California, our understanding of earthquakes changes over time.
Earlier this month, California released some revised Alquist-Priolo Earthquake Zone maps. Alquist-Priolo maps show the location of surface traces of known faults. California requires special studies to build in these areas and in some cases, prohibits construction, which can affect the value of properties for development and the ability to rebuild existing properties if there’s a fire. Our understanding of building methods and materials also develops over time. All of which is a long-winded way of saying that the outcome of PML assessments can change, so it may not be a good idea to rely on older reports to evaluate seismic risk. Now I’d like to turn things over to Jay Kumar. Jay’s our Director of Structural Engineering located in Southern California. I’ve asked him to talk about retrofit ordinances that aren’t addressed in the PML standards. Jay?
Jay: Hi, I’m Jay Kumar and I’m the Technical Director for our Seismic and Structural Practice on the west coast. I manage a team of engineers in our San Francisco, Pasadena and San Diego offices. Our structural practice carries engineering licenses in areas of high seismicity California, Washington, Oregon, Nevada, Utah, and Tennessee. I live and work in Los Angeles, so I spend more time thinking about the LA retrofit ordinance, partly because the process is still just waking up here.
The city of Los Angeles is in the process of sending out 13,000 notifications on wood frame soft story buildings and probably another 2,000 larger concrete buildings in the very near future. The current city language is, “Retrofit the building or demolish it.” And the political motivations appear to be both a combination of life safety preservation and economic resiliency. If the buildings collapse, people leave and the economy takes years to recover. So, every day here, we field dozens of calls with building owners trying to come to grips with what they’re supposed to do to comply and it’s not always an easy process. And navigating the requirements in the city of Los Angeles can be challenging.
So, the cities around California have been implementing ordinances requiring structural strengthening of buildings. These ordinances target the weakest buildings most likely to be damaged or collapsed during ground shaking. In both San Francisco and LA, buildings with soft story or weak story conditions have been recently targeted. Now, these are buildings typically with large areas of ground floor parking or tuck under parking. And these are buildings that have a history of collapsing during both the 1989 Loma Prieta and the 1994 Northridge earthquakes. Unreinforced masonry buildings were previously required by the state of California tube and retrofit in the late 1980s and early 1990s.
The San Francisco program was signed into law in 2013 and this really started the domino effect. It was signed on the anniversary of the 1906 earthquake. The goal of the ordinance was simple. It’s to improve the resiliency of the housing stock by retrofitting older wood frame buildings. So, as we saw in the 1989 Loma Prieta earthquake, these structures were particularly vulnerable to damage, especially those that were built on areas with soft soil conditions. In San Francisco, screening forms were submitted to all the owners back in September of 2013. And the screening forms are completed by a licensed design professional. There the engineer determines if in fact, a soft story condition exists or if there had been a previous retrofit maybe done under a voluntary program.
And from there, buildings were classified into one of four compliance tiers and required to have the work completed on the timeline shown here. More recently, in November of 2015, Los Angeles enacted what is considered to be the country’s most stringent seismic retrofit ordinance. And the buildings affected by this ordinance are wood-framed, multifamily buildings that contain large areas of tuck-under parking, with a stability weakness at the ground floor level, commonly known as the soft story or weak story condition. Affected buildings are those with four or more units built before 1978.
The second type of building are non-ductile concrete buildings. Now these are structures with concrete floors, concrete roof, supported on concrete columns and concrete walls. And this includes all concrete buildings built to codes prior to the 1976 Uniform Building Code. So, the County of Los Angeles is huge, it contains 88 cities. So, it’s important to note that only buildings within the city limits are those affected. Courtesy notices were mailed out to all the building owners informing them that their buildings are on this retrofit list that’s held by the city. Now we expect the orders to comply, this is the secondary notice, to take about another year to all be sent out to the owners. And this is coming out at a rate of about 1,000 per month where the high hazard or Priority 1 buildings will be targeted first. And the first wave should be reaching building owners right about now.
You can actually go online to see if the building is on the list and it can be accessed through the LADBS website. Now, the obligation to comply with the ordinance is not actually triggered until the owner has received the order to comply or if it’s been sent. And this starts the clock ticking. And it essentially provides a seven-year timeline, but we think that seven-year timeline is going to create a very false sense of security. For the clients that we are working with that are already out in front of the ordinance and in construction, it typically takes about six months to a year to get into construction. And we expect the process to streamline some as the city adapts to the process but right now it’s still fairly slow.
Since November, only about 60 building permits have been issued since November. So, there’s still quite a ways to go and we anticipate some bottleneck coming in the future. Now, if you’re starting the process at Year 5, you may be too late and we believe that that could become problematic for many building owners. Those familiar with the San Francisco ordinance may notice a couple of differences right away. Now in Los Angeles, two-story buildings are included. They were excluded in San Francisco. The primary design code is Division 93 and that’s an LA standard. The buildings on the list here are pre-identified by the city. And there are a number of different types of retrofit solutions that are discouraged.
So many people want to know “How long does this take?”. And our experience is, in addition to your building permit, there’s a very lengthy housing submittal called tenant habitability plan. And the building permit won’t actually be issued until this housing clearance is obtained. And the occupants must be notified 25 days in advance of construction. It also includes things like lead and asbestos testing, so there are a number of hurdles that can get in the way. So, after hiring your engineering contractor, plan on about four to six months before you get into construction.
The other question that comes up all the time, and it’s a great question, is how much is it going to cost. Now, here’s an example where both of these buildings are on the LA City retrofit list, they both require seismic upgrades. The building on the left has two units over an unstable configuration. The building on the right has 50 units over tuck under parking. Now, both buildings have about the same number of units, but you can imagine the costs are going to be substantially different.
And construction costs really depend on the building configuration. So, a good rule of thumb is assume the steel moment frame, one frame for about every five parking spaces. And that turns out to be about $5,000 per under parking space for construction. Soft costs, your engineering and design, typically range about 10 to 15% of the construction costs. And that’s partly due to just the many ancillary costs with the permitting process. So, the city has cast a very wide net and there are a number of buildings that do not require upgrades. In most cases, the Department of Building and Safety does not have the blueprints for these buildings. And so, they had to make their assessments using Google Maps, aerial images and in some cases, actually physically walking down streets.
But many of these conditions are at the backs of buildings and it can be hard to tell what’s actually there. There are limits to what you can see in Google Earth. So, in many cases, hiring an engineer to appeal the order is worth the money. Typically to determine that your building is okay, it involves doing a couple of things to really check to see what you have there. It often means removing finishes, looking for the building plans and doing very detailed site inspections with your engineer. So, in summary, there are really three options you can take can here. You can appeal, you can comply or you can ignore.
And the do-nothing option is always a reasonable option. However, most building owners should be aware that there is case law in California where building owners chose to ignore past their completion deadlines, they had time extensions but were later found liable for damages related to the deaths of two building occupants, resulting from the San Simeon earthquake. So, a few pro tips are don’t over strengthen the ground floor and this is worth discussing with your engineer. If you do so, you can drive the damage into the second or higher floors.
Now, this is often overlooked by most design engineers. If you’re over-strengthening, you’re also simply buying too much retrofit. And the other item is, it’s important to communicate with your design engineer about mandatory minimums or voluntary components. The retrofit ordinance only focuses on the primary deficiencies in the buildings and it doesn’t really necessarily improve the entire ground floor. So, understanding the difference of that performance is very important. Now, voluntary and mandatory components should always be clearly labeled in the design drawings and referenced in the contract, so that everybody knows what they’re getting.
The other is soft story design is really kind of a structural engineering niche. Most large design firms are focused on new construction and large development and it usually takes a middle-sized or right-sized engineering company to fit the bill perfectly. The other is watch for technical amendments. The city is publishing updates all of the time, and make sure your design professional is aware of these. They are not announced, they simply show up quietly on the LA City websites.
The second type of building that is affected by the ordinance in Los Angeles are the non-ductile concrete buildings. After this 1971 San Fernando or Sylmar earthquake, the concrete design codes required that concrete be designed to fail in a ductile and not brittle manner. These two photographs are examples of non-ductile concrete behavior. Modern concrete design seeks to avoid this type of failure from occurring so that buildings can be evacuated safely after a major earthquake.
The non-ductile concrete retrofits, they’re more complex, they’re much more expensive and the timeline is quite a bit longer to achieve compliance. So currently we’re looking at a quarter of the century. There’s a 25-year time period in which to achieve compliance. The technical amendments to this component of the ordinance are also still in the works and many questions still need to be answered before it makes sense to pursue the retrofit in the most economical way.
This process can be expensive with construction costs ranging into several million dollars. We see costs that vary anywhere from $20 per square foot to $100 per square foot. And this often is a function of what other kind of building renovations are pursued at the same time as the retrofit. Now, we expect that there will be improved guidance on approved methodologies in the near future. The existing buildings committee and SEAOSC, the Structural Engineers Association of Southern California are still flushing out many of the technical details of the ordinance and we expect many changes to occur.
There are a number of different options to comply and different ways to address seismic vulnerabilities, so I’m going to talk through just a few as examples here. And the best option is always determined based on building type, owner objectives, budget and available financing and city requirements. And you know, your costs often depend on the complexity of the project. Kind of the basic approach for the wood frame buildings is always to look for more economical solutions. That’s typically looking for plywood shear walls and looking for the addition of plywood where appropriate. And this involves strengthening the ground floor or whatever open line you have on the building using plywood and these costs tend to be on the lower end of the spectrum.
The second most common type of solution in the parking areas are what we refer to as cantilever columns. These are steel wide flange columns, independent, they’re set into large concrete foundations below grade and they provide improved stability compared to the existing slender columns already in place. The most common type of retrofit that you’re going to see in Los Angeles is something that’s referred to as a steel moment resisting frame. And this is a combination of a beam with columns with welded fixed connections also set into a fairly aggressive foundation below grade. And this is generally the most cost-effective way to approach it and the preferred option because it does not impede or obstruct the parking areas.
So, these are really the three most common approaches that you’ll see in LA for strengthening wood frame buildings. And with that, I’d like to turn it back over to Joe to discuss some of the value of seismic retrofit in the context of insurance.
Joe: Thanks, Jay. Jay and I did this webinar for a bunch of building owners right after the ordinance came out and we had an insurance professional that did it with us. I’m going to try to kind of do a real short version of a case study presented. We had a site that we looked at and it required $150,000 retrofit. And of course, when the City sends a building owner a letter saying you must spend $150,000, building owners don’t like it at all. But we try to make lemonade out of lemons here and say, “Hey look, maybe it actually saves you some money in a way. The difference in insurance costs was $15,000 per year and so if you cap that at five cap which is common in Los Angeles, that’d be a $300,000 value and so in a way it does have a return on investment.”
Now, of course a lot of people aren’t insuring at all, they’re running naked on insurance and that has its own risk to it, right? So, here you are living with that risk and in the state of the world, you’ll see the expense of that. We also spend a lot of time on policy for lenders. Now, I doubt a lot of people on the call write policy for lenders but I’m going to hit a couple of points because we’re real active in doing that. We helped both Fannie and Freddie with their policy and we’ve written policy for maybe two dozen regional lenders. And so, give us a call both Bill and I are good resources on policy. Because you know, Bill has done it in the big leagues for Wells Fargo and I’ve helped a lot of lenders too.
When we do policies for lenders we spend half the time with a simple question, “When should I order?” The regional lenders say, “Well geez, I don’t want to order on every job because every deal I have because it’ll put too much cost and make us less competitive. Well I’ll only do it in certain locations where the ground shaking’s high, you know, I’ll only do it when the building’s a certain age or maybe I’ll only do it when the size of the loan is over a certain amount.” And we just help people build those triggers, it’s really not that complicated, but we can give you advice and kind of what the guys across the street are doing.
We also help them with a little section in their policy on vendor qualifications. You should hire a licensed engineer to do this work. I also think that you should ask your engineer to carry $10 million or more in insurance. Our firm does like 40,000 transactions per year nationwide, environmental and PCAs. We do have a claim here and there on our insurance. But when you’re doing the environmental practice, you’re going to have a claim, let’s say you have one claim you know, every other year or something. Okay, that’s fine for your insurance of let’s say, you have one or two million-dollar insurance then you should be covered.
But what’s going to happen in your seismic practice, your 20 years of practice, all of those claims are going to happen in the same year. And therefore, you should ask your engineer to have more insurance. You know, one of the key things in ASTM is this level 1 or level 0 or level 2, even. And you should specify, ask for level 1. That means an engineer that goes to the site and that is most of our practice. We do do level zero, level zeros are not an outrageous thing to do. And we do do more sophisticated level 2 or 3 studies on occasion too. But a couple more quick points for seismic policies is credit side ordering is always important because if the same human being is asking the engineers the same question, you’re going to get a lot more consistency in terms of the answer.
Everything we do in a PML is really more of a statistical exercising. It doesn’t consider the ordinances. You might want to ask your engineer specifically to address the ordinances. And then a pet peeve of mine as a person that’s doing, on occasion, peer review of my competitors work, is it’s nice to show the math and to make it somewhat transparent and peer-reviewable. The PML’s a statistical model and doesn’t really talk about the ordinance. You may have properties in your portfolio that are getting hit with this ordinance Mr. Lender. And what are you going to do for these building owners? They’re getting a letter from the City saying they need $150, $250, $300,000 worth of work. Perhaps you could lend them the money to do the work. That would be great.
And I realize that there’s loan covenants, that don’t make this super easy. But I still think, if there is senior credit guys on this call that you should think about how to enable this because there’s other mechanisms available to borrowers like PACE-financing where they put the debt on their tax bill and that actually would step in front of the first mortgage. I don’t think lenders prefer that, they would rather lend the money themselves. Here’s just a quick list of six things that lenders can do, when underwriting an asset that they know is on this list.
Both Fannie and Freddie in the last, let’s see three years, Freddie more recently than Fannie, completely redid their policy and we participated in both processes. They both have more rigorous requirements and so to some extent they were in front of ASTM in terms of requiring qualified people to do these PML reports. And they both recognize USGS maps for as to when a seismic reporter is triggered instead of the UBC. The CMBS lenders, the CREF C standard reps and warranties is kind of vague, it really refers to a PML. It says that if the PML is over 20% and earthquake insurance is provided. It doesn’t really say specifically that there’s that you know, how to approach it and it doesn’t refer to ASTM specifically.
ASTM refers to CMBS though. ASTM E2557 specifically state that the PML for the purposes of CMBS financing should be considered the SEL. So, what’s going be the impact of the new ASTM standards is important. It’s just out, I mean, we are three weeks into this, but I believe it’ll create more consistency in the field. And I talked to a client that there could be higher prices and they said, “You know what, it’s worth it. I would love to see more consistency in the practice of PMLs.” So, I think people are going to want to order these things from engineering firms more often than just have the environmental guy kind of toss it on as an extra.
There’s probably a shortage of qualified staff. I mean, look, I’ve been aggressively hiring structural engineers for three years and I can tell you that the unemployment rate for structural engineers is zero. So, there’ll be a little pressure on both turn around and fees. About half the people on the call are probably from the equity side, you’re building owners. You may order an ASTM level 1 PML which is a great value. It’s a statistical model. You may consider a more detailed study like a level 2 or ask for an ASCE 41 assessment. The difference is, is then the engineer will do load calculations. We’ll check the element in the load path and make sure that each of these structural elements will perform and it’s just better than the statistical models.
A lot of guys at the end of the day do want a number, a PML number and so sometimes people will order and we will produce an ASTM PML in concurrent with an ASCE 41 study. The last bit of information is, I think you’ve heard people before me, emphasize the importance of plans when we’re doing a PML. Borrowers are used to getting the request for plans but not necessarily providing them. The fact is for a borrower, to produce binding structural plans and giving them to us is very difficult. And one way that you might handle that and it’s a really a fair way to do it, is say “Hey look, the PML is going to cost X dollars. But if you are able to provide plans, the engineer can do it at a $500 discount,” or something like that.
And that really is appropriate because when I get plans and I happen to be doing a PCA too, I could have my architect do the site inspection if I have the structural plans that my engineer can look at, valid structural plans. That really does save man hours, and it’s commensurate to a decent amount of money.
All right, well, I think that brings us to close. Thank you very much. Feel free to email any of the three of us with questions. You can see our email here. The partneresi.com website, there’s a lot of resources on seismic. If you go to the resource section of our website, you will see other seismic resources and other webinars available on demand as well as webinars on other subjects, so please consider that. Also look out for our blog on this subject, on globest.com. Thank you guys very much for your time and feel free to reach out if there’s any way we can help.