The Central European University and the Center for Climate Change and Sustainable Energy Policy (3CSEP) is inquiring about the "lock-in" effect on facility energy design decisions.  One variable of this effect is duration between renovations. These are some of my notes.  

Introduction to Lock-In

The end of a construction project is not exactly a good time to discuss further facility changes.  Owners tend to move on, not considering changes again for a very long time.  We accept the deficits of the new situation, "locking" them into place until the next renovation.  Maybe the carpet wasn't quite what we wanted.  Next time.    The roof shape, or window arrangement wasn't quite right.  Next time.   Changing it now that we've moved in would take too much ... of everything.  Next time.  We seem to like our environments stable.   As it turns out, there have been a series of lively studies on the subject of irreversible decisions.  We say we can change at any time.  Indeed, technically true.  But reality often intervenes. This is a constant concern for those of us who build capital intensive projects.  We know the consequences of our decisions will be around for a long time. So we strive to get them right.  One of our challenges is the scope of our decisions.   One architectural design reference puts the number of decisions at 133 in 20 major categories (pages 36 & 37) .  Energy is among them, but it competes with equally important, and not so important decisions.  The effect is that the cognitive load of decision makers is prioritized, shared and phased throughout the design process. So how do we ensure that energy consumption decisions are at the top of the list?  One viewpoint is to advise decision makers the risks of not addressing the issue.  One of those risks is locking in poor energy performance which contributes to climate change.

Lock-In, Facilities Energy Consumption & Climate Change

The climate and energy impact of building design is this:  if you do a "standard renovation" - large enough to warrant a building permit - the building code may require you to enact measures which may reduce your power consumption by as much as 15%.  A "moderate retrofit" may reduce your consumption up to 30%. But what if much larger reductions could be had? Imagine this scenario playing out across 5% of buildings every year.  On the scale of a city, we can get a sense of when energy efficiency decisions would yield results. If enough cities do this, we can summarize the effects across countries.  Because buildings consume half of our energy, that becomes significant for energy and climate policy.   One conclusion:  a "good enough" renovation delays carbon emission reductions necessary to achieve our climate change goals.

A Different Choice

Let's say a year goes by.  You discover that for a modest amount more, you could have reduced your energy consumption 80%.   What would you do?  Have it done right away, or wait for the next renovation cycle? Traditional business practice - and human nature - holds that you would wait until the next renovation.  So the question of CEU is: when would that be? How long for us to enjoy the results of you using a lot less energy? This does not turn out to be a straightforward.

Method

To answer the question for average time between renovations, I made two key assumptions: 1.  Such a number exists. 2.  The number includes "commercial" structures: commercial defined to include government facilities. I canvased resources here in my office - including a decades old textbook on engineering economics.  I consulted my network for the information.  I also conducted internet searches for information from government and the engineering, finance and insurance industries.  As a last resort, I posted a request for information on LinkedIn building design and management group forums.  I can't claim academic rigor on the research thus far.  What I can claim is staking out the corners of the research. The general conclusion can be summed as, "it depends."  As one colleague put it, "it's an impossibly broad question."  Well.  This is Landwave.  Impossibly broad is for breakfast.

Driving Factors

The following could be considered possible drivers that compel owners to change their buildings on the scale of an energy retrofit.

FINANCIAL ANALYSIS

A frequent practice for building owners is a cost-benefit analysis for various courses of action under consideration.  In 2011, the US Pacific Northwest National Lab produced a series of Advanced Retrofit Guides for a variety of building types.  Among the extensive data presented, the Office Guide uses a 20 year analysis period.  This is reasonable, but subject to extreme variations. I would think their work would be among the most reliable in terms of academic rigor.  But I wouldn't count me as a good judge. See page 118 (print page numbering)/122 (PDF page), http://www.pnnl.gov/main/publications/external/technical_reports/PNNL-20761.pdf

FACILITY DURABILITY

Another possible approach to determining time between renovations: examine average durability of subsystems.  Material physics, environmental factors, installation quality and maintenance regimens govern the useful life of facility components.  Failure of any one of these systems would not necessarily justify a retrofit.  But every failure brings up the question and the opportunity.  I've been involved in many a building renovation due to a single system failure. All of these durations are from my experience.  It is entirely possible to retrieve a great deal of citations from around the building industry to support these ranges.  A hefty task, but feasible.

Concrete can last for centuries without serious maintenance when installed properly.  What degrades is surface appearance.  There can also be localized structural failures due to the consequences of spalling. Neither event tends to trigger wholesale retrofit.

Masonry, like concrete, is incredibly durable.  Skin failures are rare and often due to neglect. Inspections every 5-10 years for mortar failure are the common practice.  Mortar pointing is recommended maintenance and not terribly expensive.

Wood, in it's raw state is highly vulnerable to deterioration from exposure to weather.  The typical response is regular application of coatings.  Plenty of examples abound of centuries old installations that survive with proper maintenance.  Repainting happens in 5 to ten year cycles, though plenty go longer.

Exterior Insulation Finish Systems  (EIFS)  are probably the most controversial and delicate of building skins.  EIFS requires regular inspection and care.  To meet owners demands for the most carefree installation possible, strict manufacturers inspections are required to guarantee an installation. And that duration is usually quite short - 5 to 10 years, though some swear longer.

Glazing, like concrete and masonry, can theoretically last centuries.  The point of failure for glazing is typically frames and attachments. Caulks and sealants are the first line of defense against frame degradation.

Caulks & Sealants effectively stop moisture and vapor penetration at adjoining materials.  Re-inspection and re-pointing is generally done every 4-5 years, with expected overall replacements to be had every 20 years or so for well maintained and installed systems.  10 otherwise.

Metals wear based on type, thickness, surface finish, installation quality and environment.  The variety of finishes and metal types available for cladding are voluminous.  Figure a mid quality, well installed and maintained metal facade element could last up to 30 to 40 years.

Roofing is the most reliable of all systems to estimate life cycles on.  Roofing systems carry warranties of 20, 25 and 30 years for both residential and commercial installations. Many last well beyond those warranties, but those time horizons are the trigger points for a shift in liability.  In practice, a roof has to be in bad shape, an additional 5-10 years beyond warranty, before facility operators consider replacement.

Mechanical Systems primary components can last 15-20 years with proper maintenance.  Small scale installations are largely ignored by owners until a failure forces their hand.

Plumbing Systems  in theory can last 20-40 years without significant deterioration.  Fixtures change out every 15-20 years due to failures. Equipment such as water heaters have been anecdotally engineered to precise failure at 20 years.

All in all, this method of estimating time horizons does not yield a concise number.

INSURANCE ACTUARIALS

I have often assumed that the insurance industry might maintain a collection of deterioration information to analyze their risk across differing kinds of building types.   This evidence might prove a useful basis for underwriting policies. Not being a statistician, I would think a deterioration data set from insurance claims would provide a statistically useful data set for this type of analysis. Apparently that's not the case.  The insurance industry does not seem to have leagues of engineers and analysts in the background keeping track of the durability of building components. The insurance industry therefore would have no independent basis for determining their risk when covering, say, a high rise building. They calculate risk on replacement value, not deterioration.

TAXES

Another possible driver in the US might be the tax code.  Commercial property depreciation is required in business tax filings and spans 30-40 years depending on building type.  Rates of depreciation are based on straight line curves, much like amortization. That is, take the expected loss in value, divide by the period in question and apply the result each year. Where these time spans originate is not clear. Depreciation is incredibly difficult to grasp in buildings. The basic concept is that an asset purchased for a business will lose its resale value over time. Manufacturing equipment and vehicles are the easy examples.  A car bought today is worth less in five years. However buildings don't depreciate uniformly. Nor does deterioration of components uniformly impact resale value in robust markets. While building condition is accounted for, it does not tend to match depreciation. In a hot real estate market,  I can sell a dump of a building for more than I bought it for.  Conversely, during a real estate downturn, a building kept in "tip-top" shape with a sensible maintenance regimen may suffer a loss. The advantage of depreciation is that I can claim a loss on my income that I might not actually have. Not exactly a driver. This and the insurance question was verified through conversations with two seasoned colleagues with deep insurance industry experience.

BUILDING PERMIT HISTORIES

A suggestion made by colleagues would be to canvas the building permits of jurisdictions for changes occurring at specific addresses.  While that might give us insight into the frequency of change, it would not control for rationale.  Nor would I expect to see a consistent frequency of replacement.   A deeper analysis of building types may be obtained as Building Type is regulated in building codes.  What it won't say is the details of building skins. A constraint to consider is that jurisdiction records are inconsistently maintained.  Digital records may only extend back twenty years or so for the jurisdictions that have implemented digital record keeping.  And as you may be aware, "digital" does not equal "consistent." Hard copy search would be required otherwise. It should be taken into account that jurisdictions don't take consistent care of those records. Such a dataset does not seem to exist.

COMMERCIAL REAL ESTATE LOAN PERIODS

Typical duration of commercial real estate loans is 20 to 30 years.  Loan events do not uniformly correspond to a renovation. See the Comptrollers Handbook from the US Department of Treasury on commercial real estate lending, page 38 printed, PDF page 42: http://www.occ.gov/publications/publications-by-type/comptrollers-handbook/cre.pdf   This handbook covers bank examination  practices for commercial real estate lenders.

TENANT/OCCUPANCY CHANGE

Commercial leases in the US are on average 10 years.  Often the building shell and equipment is not a point of contention in the lease.  Tenants typically prioritize usable space in a good location for rents that work within their business models.  Though there is a surge in leases in green buildings, it's not clear what is driving that surge.  Is it simply pent up demand, or is there a clear advantage in the minds of tenants to be in green buildings? Anecdotally, I recently inquired at a Tesla dealer about the "green" features of their facility.  With some embarrassment they confessed there were none.  They had simply leased the most convenient facility - smack in the middle of their clientele.

Interim Conclusions

While not exactly the hard number we were seeking, my efforts thus far may at least provide the basis for understanding the difficulties in finding such a  number. My effort also may lead to approaching the argument differently.  Perhaps instead of citing conclusive statistics on actual durations, approach it by stating "common analytical practice" is a range of 20 to 30 years.  Comparative analysis can then be carried out over that range to see what conclusions can be drawn. Stand by for other parts of this analysis.