Vitruvius has something interesting to say at De architectura 2.8.8:Footnote 1
Non enim quae sunt e molli caementoFootnote 2 subtili facie venustatis, non eae possunt esse in vetustate non ruinosae. itaque cum arbitrio communium parietum sumuntur, non aestimant eos quanti facti fuerint, sed cum ex tabulis inveniunt eorum locationes, pretia praeteritorum annorum singulorum deducunt octogesimas et ita – ex reliqua summa parte reddi pro his parietibus – sententiam pronuntiant eos non posse plus quam annos LXXX durare.
Those structures made of soft rubble, for all their subtle attractiveness, are not the ones that will resist ruin as time passes. And thus when assessors are appointed to evaluate party walls, they never assess soft rubble walls according to their initial cost, but rather, when they look at the price recorded in the original contracts, they deduct 1/80th of that sum for each subsequent year, and the remaining amount is fixed as the current value of the walls. They have rendered judgement, in effect, that such walls cannot last more than 80 years.
By contrast, mud-brick walls were not depreciated at all if they were standing when assessed (Vitr. 2.8.9). The calculation performed for concrete walls demonstrates clear understanding of what we call ‘depreciation’ or ‘amortization’ in ancient Roman thought. It appears to have been overlooked to date.Footnote 3
Depreciation is an economic concept of some sophistication,Footnote 4 and its presence in late Republican Rome is significant for any discussion of economic and moral values at that crucial time in ancient history.
When Vitruvius was writing, in the second half of the first century b.c.,Footnote 5 houses were frequently collapsing, burning, being sold on, demolished and rebuilt.Footnote 6 Shocking stories of sudden tenement collapse may have motivated the arbitrators' allocation of a service life to concrete walls: when buildings collapsed, subsequent investigation into their construction would have revealed their age. Eighty years is likely to have been a convenient approximation based on whatever sample the arbitrators in question could find at the time they framed this accounting procedure. Building concrete walls had begun about 200 b.c. and is first mentioned c. 160 b.c.Footnote 7 As with any new technology, early efforts probably suffered from numerous defects, and it would have taken years or decades for concrete technology to stabilize and standardize. In due course the Romans really honed this technology: the best ancient concrete structures still survive thousands of years after their production, even exposed to the elements.Footnote 8 But early concrete was not so good, and not all concrete produced was of the same standard. That constructed with less care, of less well-prepared lime, mixed in less than perfect proportions, or in an unstable environment, has since disappeared.Footnote 9 The arbitrators allocated 80 years for the service life of concrete walls on the basis of their experiences and investigations, sometime before Vitruvius was writing (say about 30 b.c. for present purposes), so the tenement buildings whose early concrete walls had failed so spectacularly as to initiate awareness of the concept of depreciation, would have been constructed c. 110 b.c. or earlier.
Should we assume that, when legislating for the future, the arbitrators were aiming for worst-case scenario, the earliest age at which a concrete wall might collapse, or an average lifespan of such walls? The answer to that question largely depends on whether the arbitrators were concerned about the health and safety of the occupants of the buildings, or the balance sheets of the owners of the buildings: who were the consumers of these depreciation calculations? At this time and place in history, Late Republican Rome, the answer is probably the owners and potential owners of tenement buildings, who would want evaluations of party walls as part and parcel of estimations of the value of their estate, or if the building was needed as security or suchlike. Moreover, it is the owners, not the tenants, who would have possessed the construction contracts which recorded the original cost, with which the calculation necessarily began.Footnote 10
Maintenance and repair of buildings was a major issue in the second and first centuries b.c., and the quality of the concrete used during the early years of its employment perhaps explains why.Footnote 11 It was not just tenement blocks that were affected. Extensive repair of public buildings was needed in Rome when Augustus started his programme of restoration.Footnote 12 His important civic reforms included reorganization of the administration of public building,Footnote 13 institution of the cohortes vigilum, and some specific statutes, including limitation of the height of buildings to less than 70 feet and the width of party walls to not more than one and a half feet.Footnote 14 Private houses suddenly falling down, as reported by Strabo, and the simultaneous review of public building theory and practice, offer a coherent historical context for the inclusion of a depreciation clause in Roman contracts on party walls.
How sophisticated the Roman arbitrators' understanding of depreciation was can be appreciated through comparison with other times and places. The next reference to depreciation in historical sources known to me was written about sixteen centuries later, in Britain, and it reveals a simpler understanding of the concept than that preserved in Vitruvius. Robert Loder's farm accounts for 1612 appear to include a sum for depreciation of horses and their gear: ‘Other charges which I was at aboutes my horses … the wearing out of my horses; with the wearing of other thinges, as followeth … Item I reckon my horses were the worse in the abovsaid time by iijl; Item I judge, that the wearing of other thinges, as forehorse halter, ring'd halters, collar halters, skipe, chafe seve, shovel, crippers, fetters, basten ropes, my plow, harrowes, etc. cost me xvs iiijd’ (1612). In the following year he charged for ‘horses the worse by wearing out, and in wearing other thinges: as for horse halter’ [the list is repeated] ‘… in wearing out of the sayd horse which come to I iudge, vs’, and ‘Item for the wearing of [that horse] out some in the time I mowe, viijs iiijd’.Footnote 15 These charges are not expressed as a fraction of the anticipated service life of a horse, or of the estimated value of a horse, nor is there any attempt to relate the sum charged for depreciation of four horses in the next year to that charged for the four (or additionally for one that had performed extra work) in 1612. Further, Loder did not make similar allowances in eight out of the ten years for which accounts survive; normally he only charged his feed costs. His accounting practice, in short, was inconsistent, and his 1612 and 1613 behaviour appears to have been a short-lived awareness of depreciation (prompted perhaps by his recent purchase of these horses or, and this seems less likely given Loder's obvious concern for his horses' well-being, by their palpable exhaustion). In contrast, the more refined calculation of depreciation by the arbitrators mentioned by Vitruvius appears to have been routine.
Another example of depreciation dates to 1675, when Stephen Monteage devalued six horses by a sum that equates to one seventh each for ‘their use and impairing’. Mason thought it likely that those ‘regularly engaged in business’ have always understood ‘the fundamental facts of depreciation’ and that, were the records available, we would ‘find evidence of some understanding of the phenomenon of depreciation as far back in history as the origins of written records of business affairs’.Footnote 16 But I find no such evidence in Roman agricultural treatises, which have survived in relatively good quantity, or in the Heroninos archive.Footnote 17 Neither Vitruvius' contemporary Varro, nor Columella, about 50 years later, talk of depreciation with respect, for example, to animals, farm equipment or farm buildings, in the same clear way as Vitruvius does of concrete walls in 2.8.8; when talking of carting goods around the estate, for example, Varro notes a possible choice between donkey, cow and mule, and observes only that a donkey is cheaper to feed than a cow, but a cow is more productive (fructuosior, Rust. 1.20.4). Macve pointed out that thirteenth- and fourteenth-century writers ‘were able to recognize, at least implicitly, the effect of depreciation on investment choices – e.g. in comparing horses and oxen they recognized that a horse's carcass was worth little, while an ox could be fattened for meat’.Footnote 18 But the fact remains that evidence for real understanding of the concept of depreciation is wanting through most centuries in most places. I have not been able to find evidence for it in Italian accounting up to and including Pacioli, who knew Leon Battista Alberti, who studied Vitruvius intensively and who thereby opens the door to the possibility of transmission of the concept directly from Vitruvius to Renaissance Italy.Footnote 19 In 1764 John Smeaton, writing an expense report for maintaining the Forth to Clyde canal and apparently without any link to Vitruvius, supposed that ‘in 20 years’ time many of the locks will want new gates, all of which will gradually fail in a few years after. I therefore suppose them all made at the end of 20 years and estimated the expense accordingly'.Footnote 20 American accountants did not get interested in the concept until the 1830s, when large capital investments were being made in the railway industry and stockholders required their business managers to account for the assets. As late as 1912, depreciation still received ‘little or no systematic attention’.Footnote 21
An essential feature of the method of calculating depreciation reported by Vitruvius is that depreciation is uniform over the life of the asset. It follows a straight-line pattern, so that depreciation in year 1 is the same as depreciation in year 2 and so on. As a verbal formula, this is:
The current value of the asset = the original cost less the depreciation to date, where the depreciation to date = a fraction × the original cost × the age.
The fraction in this case, concrete walls, is 1/80th. In general, the precise fraction is chosen to reflect the anticipated life of the asset, and may be rather arbitrary.Footnote 22
The accounting method described by Vitruvius is in essence the same as accounting procedures for depreciation used (with many added refinements, of course) in the developed world until the 1990s, when the U.S. Bureau of Economic Analysis switched to a geometric pattern for most assets. Using a geometric pattern, depreciation is reckoned a little higher in the early years of an asset's life than in the later years.Footnote 23 Ancient historians and classical archaeologists may be amused to learn that the service life expected of new residential structures in the United States is 80 years,Footnote 24 the same as concrete walls in Vitruvius' time.
Another perspective on how advanced the Romans' understanding was of the concept of depreciation emerges from consideration of the particular asset being depreciated in Vitruvius' account: walls. Unlike, say, horses working on a farm or lock gates moving back and forth to retain and release water, in which cases wear and tear is ongoing, cumulative and usually visible, walls do not significantly decline in useful service gradually but, like lightbulbs, typically work without problems until they fail suddenly and completely.Footnote 25 It is quite remarkable that in such circumstances, in the first century b.c, the value of the anticipated loss was being averaged out over the years of expected service.Footnote 26
