Like a War Zone

As catastrophic as Hurricane Katrina, centered on the industrial north—even with serious undercounting and omissions, that’s what $$ numbers in official reports document for the March 1913 storm system and resulting natural disaster
 

Like Dresden after the bombing in World War II. 

That was my instant thought the first time I beheld—and gasped at—a Library of Congress panoramic image of still-smoking ruins amid rubble left of downtown Dayton, Ohio, taken immediately after the Great Easter 1913 floodwaters had receded, and while fires from ruptured gas mains were still smoldering. Another stunning vista 

Dayton, Ohio, in smoldering ruins after the March 1913 flood. Library of Congress. See also this 5:37-minute silent newsreel footage in this very area of Dayton on YouTube, especially between minutes 3:00 and 4:00 panning the smoldering ruins. Like Dresden, indeed.

shows acres of suburban houses submerged up to their eaves. Official maps documented that 42 percent—7.2 square miles—of this fast-growing industrial city of 

Dayton homes submerged up to their eaves in March 1913. Dayton Metro Library

about 126,000 was inundated and swept by mighty currents. Yet the business journals Dun’s Review and Bradstreet’s in their April issues covering the flooding initially estimated only $30 million in damages—way under the figure exceeding $73 million later documented by the Dayton Citizens’ Relief Committee’s detailed inspection of the worst devastation (Horton and Jackson, page 53).

Seven hundred miles west, photographs of the southwestern outskirts and downtown Omaha, Nebraska, taken after the EF-4 Great Easter tornado destroyed a swath blocks wide and miles long, resembled Dayton’s rubble minus the mud. Yet tallies for damages in Nebraska, Iowa, and Missouri from the family of 10 violent Easter

Omaha Furniture Manufacturing Co, in Ralston, Nebraska, after the Great Easter Omaha tornado in 1913. http://blogs.omaha.com/2012/03/21/from-the-archives-easter-tornado-of-1913/

tornadoes didn’t appear at all in the Dun’s Review special report and follow-up reports, which focused only on the flooding. To its credit, Bradstreet’s did estimate $5 million in damages for Omaha—but nothing for the rest of Nebraska, Iowa, or Missouri, despite the fact that a swath through Council Bluffs was left in ruins by another tornado in that family. Nebraska and Iowa didn’t even get mentioned in the National Weather Service’s comprehensive report Bulletin Z, which focused exclusively on the toll from flooding—even though the tornadoes were a weather disaster recognized to be part of the same storm system and were well discussed in the NWS’s regular Monthly Weather Review. Moreover, the six different authors of Bulletin Z conflict among themselves in the dollar figures they cite in various states. Yet Bulletin Z is the only official document that attempted to tally damages of the 1913 flood through the southern states as the flood crests continued to set record heights down the Mississippi River through the end of April.

Thus, as with trying to count fatalities in ‘Our National Calamity’ (see “‘Death Rode Ruthless…’”), a reader can feel hopelessly bewildered by a mass of data that is clearly incomplete and contradictory, even within individual reports. However, by assessing numbers in ALL the known official reports side by side in Excel spreadsheets, AND by making informed choices in the case of contradictions about which data to accept and which to reject based on internal and external evidence, AND by perceiving that some truly important data were omitted, a modern analyst can come closer to what may be regarded as a minimum realistic dollar figure for property damage.

So, with the caveat that because of its complexity my own analysis is still not complete, below are some key preliminary observations from my collation of all known official reports—offering numbers that are closer to reality than any published in any single official report at the time. Any future analysis will likely make the numbers grow only bigger.

First, the reports were published by different organizations: state or Federal agencies, universities, business publications, railroads, and others. Depending on the report and the concerns of the authors (see references), damages may be reported by city, county, industry, or watershed, so direct comparisons cannot be made among their component parts. Some may concentrate only on parts of states (e.g., Horton and Jackson’s U.S. Geological Survey report focused on the portions of states that constitute the Ohio River Valley). Last, they are of differing character, which affects the weight one should assign to the numbers in each. The numbers in the business journals Bradstreet’s and Dun’s Review are preliminary estimates received by reporters on the ground in the first few weeks after the tornadoes and flood, quickly rushed into print; they do not cover all cities—most importantly, they do not report figures for the hardest-hit cities because that early even estimates were not yet available. The reports assembled by the U.S. Geological Survey (Horton and Jackson) and the National Weather Service (Bulletin Z) are systematic assessments made in the months following the disaster; even so, the numbers they report depend on the willingness and ability of sources to provide the needed data, resulting in some key omissions as discussed below. 

Second, the reports focus on direct material damage to property. Thus, they do not include the cost of the catastrophe itself—the weeks of providing temporary housing and food for flood refugees, the costs borne by municipal, state, and Federal government agencies in goods and services for clean-up and sanitation, or the money received and expended by the Red Cross in relief and longer-term rehabilitation. Although I still must research this aspect further, it is starkly clear that the costs of the catastrophe itself were considerable. The Red Cross alone disbursed more than $2 million in aid; John Patterson’s National Cash Register Co. (NCR) incurred comparable expense in goods and services. Bear in mind that the Red Cross—then very small—was only in 112 of the hardest hit communities in Ohio, half a dozen counties in Indiana, and in the cities of Omaha, Nebraska, and Lower Peach Tree, Alabama (after its devastating Good Friday tornado; see “The First Punch”). There were many additional local relief funds, generally operated by local city newspapers, that were separately collected and disbursed, and which do not appear in these numbers (a future research project!). The figures below also do not include monies disbursed by state, county, or local governments for clean-up and sanitation (the Ohio State Board of Health in its reports of May and December 1913 does cite some figures for some cities, but I have not yet tallied those numbers). Nor do the figures below include concomitant costs for medical care (which I’ve never seen tallied anywhere) or appropriations for mobilizing all 8,000 troops of the Ohio National Guard. From what I have read so far, my general sense is that all these additional costs of the catastrophe—quite apart from property damage—could add a good $20 million over the entire area afflicted (from Nebraska to New York and down the Mississippi).

It does appear, however, that official reports of property damage include both public infrastructure (bridges, roads, buildings, levees) and private property (houses, farms, livestock, crops, railroads, telephone and telegraph poles and lines). In some cases they include costs due to suspension of business from destroyed infrastructure (particularly for the railroads). For buildings, the losses focused on the structures themselves. As nearly as I can tell, omitted in all reports is any consideration of 

No personal property in submerged homes or businesses, such as these shown here in Piqua, Ohio, was tallied in official damage reports--nor was its loss covered by most insurance policies. Miami Conservancy District

personal property—furniture and all other furnishings in private homes or small businesses—ruined by being submerged, even if a house or building is left standing. Not counting the loss of personal home and business property results in significant undercounting of overall damage. Today, homeowner’s insurance estimates the average value of personal property (furniture, appliances, electronics, etc., exclusive of jewelry and certain other high-value items) within your home as a percentage—commonly 50 percent, sometimes as high as 70 percent—of the dwelling structure. Even allowing for the fact that average personal wealth in today’s affluent society is far higher than it was a century ago, wholesale ruination of the contents of tens of thousands of households and businesses in 1913 could conservatively add another 10 to 40 percent in replacement costs to the figures below.

The railroads and the wireline communications (telephone and telegraph) both had multistate infrastructure that was so decimated that even with crews of (collectively) tens of thousands of men working round the clock, full service was not restored until August. After a detailed inspection of all its lines, the president, vice president, and 

Illinois River's powerful 1913 floodwaters simply swept away railroad embankments near Beardstown and twisted steel rails. Illinois State Museum Collection

chief engineer of the Baltimore & Ohio railroad estimated its own damages to be $2.5 million, two thirds of that in Ohio alone. The Pennsylvania Lines West of Pittsburgh estimated their direct losses as $3.6 million, but by September 1913 actual outlays for repairs were running about 25 percent higher than original projections. Those estimates did not include millions more in lost revenues until repairs were complete. Other railroads did not appear to publish damages, but the main author of Bulletin Z (Alfred J. Henry) tallies railroad losses as topping $15 million in all regions—the Midwest, New England, and down the Mississippi. Meantime, AT&T lost 4,000 telephone poles and Western Union 8,000 telegraph poles, and both lost well over 1,000 miles of wires—at a time when only wealthy homes had telephones. Sometimes figures for railroads and telecommunications are included in overall damage estimates in reports, and sometimes they are not. What is important is seeing that even careful estimates—such as those by the Pennsylvania Railroad—fell short of actual expenditures in restoring the system.

So, given all these caveats, what do the reports reveal , individually and collectively?

West fork of the White River partially submerged railway bridge in Gosport, Indiana. Indiana Historical Society

The business weeklies Bradstreet’s and Dun’s Review (which later merged to become Dun & Bradstreet) each quickly reported about the same level of losses across the Midwest and Northeast: $50 million for Bradstreet’s and $56 million for Dun’s. But close examination reveals that they must have had reporters in different cities, because in many cases a lack of information from one can be supplemented by data from the other, yielding a combined total for several dozen cities in half a dozen states very close to $70 million. But that figure is not definitive for two crucial reasons. First, the estimates for two of the hardest-hit Ohio cities—Dayton and Columbus—are low by factors of 2 to 5 compared with actual on-the-ground inspections and damage assessments made in succeeding weeks by official relief committees in each city. The final total for Dayton was officially determined to be $73 million (instead of $30 million the weeklies estimated) and for Columbus at $22 million (instead of a mere $4 or $5 million estimated). Worse, because some of the absolutely worst devastated cities—including Hamilton OH (whose city fathers documented a loss of some $10 million), Zanesville OH, Terre Haute IN, Indianapolis IN, Wheeling WV, and Cairo IL—did not have figures instantly ready to report in the first couple of weeks, both Bradstreet’s and Dun’s simply noted that damage was “severe” without assigning a dollar value—with the effect that any tally of damages given for the very epicenters of the worst destruction is effectively zero!! Talk about serious undercounting! 

In their U.S. Geological Survey report, authors Horton and Jackson openly acknowledge this vast undercounting of the hardest-hit areas when discussing their own methods for gathering data (pages 84-85). They sent circular letters to some 200 towns with a population of 5,000 or more, but received only 120 replies, so already they were short 40 percent of the information they sought. Moreover, they wrote, “some of the most serious losses were only vaguely expressed” and so they were not 

1913 floodwaters down the Scioto River were powerful enough to scour out the foundations of homes along Middle Hickory Street. Such flood damage was not--and still is not--covered by most homeowner's insurance policies. Library of Congress

included in their table of total estimates! Note also one key omission of their data-gathering technique: that across the Midwest in 1913—and still today!—there are many towns with populations under 5,000, yet their losses did not get counted. Across eight states (IL, IN, KY, NY, OH, PA, TN, WV) Horton and Jackson come up with a total material loss of over $191 million and conclude: “That the actual losses resulting from the flood of March-April , 1913, will greatly exceed $200,000,000 there seems to be little doubt.” Note that their concern was the flood, and so their figures also do not include the windstorm or tornadoes. And except for Tennessee, they also do not include flooding down the Mississippi.

Bulletin Z—whose various parts were compiled by six different authors, and which has significant internal contradictions—tallies $163 million for the entire flood region, including down the Mississippi, not counting the windstorm and tornadoes. Of that, it estimates $113 million was in Ohio alone (p. 47). But comparison with other data reveals that is a serious undercount—not least of all because Horton and Jackson’s independent assessment of damage in Ohio alone tops $150 million. Moreover, official on-the-ground detailed assessments by citizen relief committees conducted for just three of the hardest-hit cities—Dayton, Columbus, and Hamilton—alone total $105 million. Furthermore, as photographs document, dozens of other cities around the state—notably Chillicothe, Delaware, Tiffin, and Zanesville—suffered tremendous 

Delaware, Ohio, was left in ruins by the flooding of the Olentangy River. Library of Congress

destruction. Bulletin Z’s estimate of $113 million (or, for that matter, Horton and Jackson’s of $150 million) also contradicts another table within Bulletin Z where damages are tallied by county (pages 67-69): if one excludes the three counties having those three hardest-hit cities, the damages elsewhere around the state of Ohio top $95 million. So just in state of Ohio alone, it appears that realistic minimum numbers for material damage are $105 million for the three cities of Dayton, Columbus, and Hamilton, plus $95 million for the rest of the state—yielding a conservative total at least $200 million in property damage. That’s $200 million just for Ohio—not for the entire multistate region as Horton and Jackson projected. Considering that Governor Cox’s initial eyeball estimate for damage to Ohio was $300 million, it is clear that this former newspaper reporter with the railroad beat knew his state.

For Indiana, Horton and Jackson (USGS) tallied $18.6 million in damages, but remember that report looks only at the Ohio River Valley watershed, so may well have missed damages in northern parts of the state. One of the authors of Bulletin Z (Norquist) carefully tallied damages to Indiana and came up with a total exceeding $20 million, but he was looking only at the White River watershed. Two decades later, after the 1937 Ohio River flood, Dennis O’Harrow of the State Planning Board of Indiana in an unpublished (typescript) report comparing earlier floods, estimated Indiana’s damage in 1913 to be $25 million. Unpublished or not, that number is the closest thing we have to an assessment by an Indiana state official, and presumably includes the entire state (although O’Harrow also lists his own set of caveats and claims his number is preliminary). Personally, my big research question about Indiana’s $25 million compared to Ohio’s $200 million is: since the 1913 flood remains Indiana’s worst flood of record just as it remains Ohio’s, why is Indiana’s damage only a tenth Ohio’s when the population was 60 percent of Ohio’s? Once again, it is my hunch that $25 million is a serious undercount; nonetheless, at this point, we can state with confidence that Indiana’s property damage in 1913 was no less than $25 million from flooding alone. Furthermore, Grazulis documents another $1 million in damage from the Terre Haute tornado.

Louisville Distillery in Kentucky suffered terrible losses in the 1913 flood. http://www.huntingtonnews.net/4188

For Kentucky, Horton and Jackson assigns $2 million in damage. Bradstreet’s assigns an initial estimate of $3 million. The only official report of the 1913 flood in Kentucky is a U.S. Public Health Service report by Lumsden that does not tally property damage, but looks at the sanitation situation in just five inundated Kentucky cities along the Ohio River (Catlettsburg, Maysville, Paducah, Wyckliff, and Columbus), which were submerged anywhere from 10 percent (Wyckliff) to 90 percent (Catlettsburg). Because the rivers inland broke flood records as well, it seems valid to stick with no less than $3 million in damages to Kentucky.

So it goes also for other states, with the additional complication that data are sparse. Below is a table summarizing figures obtained when collating official reports. Note especially the figure of $8 million for damage down the Mississippi River reported in Bulletin Z (p. 42) for three river districts (Memphis, Vicksburg, and New Orleans) plus the Cumberland River in Tennessee. Note also that no report at all enumerated damage to Cairo, Illinois, whose Future City district was documented by the Red Cross as so deeply submerged that houses were floating away. Figures simply were not available.

 

Nonetheless, all reports taken together provide clear evidence that the damage to real property from the entire storm system over the entire area (Nebraska to New England and down the Mississippi, including the Good Friday and Easter tornadoes) topped a quarter-billion 1913 dollars. If personal and business property losses are calculated at a mere 10 percent of the real property, that would add another $25 million—likely more, not only because 10 percent is likely too low a multiplier, but also because many houses and businesses that were not structurally damaged still were submerged and suffered losses to the contents of the premises. If one includes the estimated cost of relief, housing and feeding the suddenly homeless, and cities-wide cleanup and sanitation efforts, the overall loss around the nation of the March-April 1913 catastrophe topped $300 million in 1913 dollars—and in all likelihood, considering the widespread undercounting, reached a third of a billion dollars.

A third of a billion dollars in 1913: What might that mean today? There are at least seven separate methods for assessing the present value of historical money (all seven are discussed at the excellent site MeasuringWorth.com by two economics professors at the University of Illinois). They are all correct in different contexts. Yet they all yield answers that differ widely [see table for 1913 below] . Since natural disaster losses pertain to damage to and rebuilding major projects such as bridges and railways, the most relevant method for this purpose seems to be the “relative share of the GDP,” which allows comparison of the cost of construction of a major project in historical times to the value in the economy at the time as a percentage of the GDP.  To compare capital losses in 1913 dollars with 2011 dollars (the latest given on Measuring Worth.com), the “relative share of the GDP” calculator multiplies 1913 dollars by a factor of 386 to reach today’s value. So $300 million to a third of a billion dollars in 1913 would translate to a range of $116 to $130 billion today. 

Indeed, a national calamity. For comparison, Superstorm Sandy that hit New York City and the Atlantic Seaboard in late 2012 was not even as costly as Hurricane Katrina, which wreaked $108 billion in damage to the Gulf Coast. http://www.cnn.com/2012/10/30/us/sandy-records However, the Great Easter 1913 storm system was as damaging as Hurricane Katrina, only even more widespread.

Several other essential points in weighing the 1913 damage: 

The flooded Cuyahoga River drove the steamship William H. Mack under Cleveland's West Third Street Bridge, levering the bridge off its piers and throwing it into the river on March 25, 1913. It also decimated all industry in The Flats  Library of Congress

First, the Great Easter 1913 storm system struck the heart of the industrial north. In 1913, three-quarters of U.S. industry was still east of the Mississippi and north of the Ohio River. The losses were suffered not only by agrarian farmers, but by the captains of industry fueling the American economy. The floodwaters quenched the fires of blast furnaces in Pennsylvania and Ohio, twisted rail lines and took down more than 500 railroad bridges, and destroyed at least 12,000 telephone and telegraph poles and their associated wires, silencing all communications between the stock exchanges in Chicago and New York City for a day and a half, and halting the US mails for 10 days. Short-term food famines afflicted New York City and other places because floodwaters in the Midwest prevented the delivery of milk, beef, and produce. 

Second, and equally sobering: the real and personal/business property damage was by and large a straight-up dead loss to both businesses and individuals. Sure, insurance existed in 1913. But like most business and homeowner policies today (2013), most insurance in 1913 did not cover groundwater or flooding! Look at your own policy: unless you live in a designated flood zone and are required to carry flood insurance—and only perhaps 20 percent of Midwesterners do carry it—you would not be insured against flood damage should a 1913-storm system recur. Bear in mind as well that one reason the 1913 flood was so deadly is that it afflicted areas not normally prone to flooding.

1913 flood hit captains of industry and their mansions. Here, a mud-covered man with shovel rests near mud-ruined furniture from this Dayton mansion. Note that these contents were ruined even though there is no visible damage to the magnificent house itself--an example of personal property loss without structural loss. Dayton Metro Library

Last, people in 1913 accepted compromises that people in 2013 would find abhorrent, or that might even now be prohibited by municipal or state ordinances: notably, continuing to live in a house that had been pushed off its foundation and filled with mud that included human excrement from flooded vaults of thousands of outdoor pit toilets. Today most such contaminated houses would likely be condemned and razed. However, today in Hamilton Ohio, according to Hamilton historian Jim Blount, when homeowners remodel Victorian homes that were submerged in 1913, they still find the walls filled with long-dried “flood mud,” which many preserve in bottles as family mementos.

©2013 Trudy E. Bell. For permission to reprint or use, contact Trudy E. Bell at t.e.bell@ieee.org

Next time: Book Report! Recent Books on the Great Easter 1913 Flood and Tornadoes

Selected References

Bulletin Z. The Floods of 1913 in the Rivers of the Ohio and Lower Mississippi Valleys. U.S. Department of Agriculture. Weather Bureau. Washington, Government Printing Office, 1913. Principal author was Alfred J. Henry, but additional reports were contributed by five other authors. 

Bybee, Hal P., and Clyde A. Malott, “The Flood of 1913 in the Lower White River Region of Indiana,” Indiana University Studies II (22): 105–223, October 1914.

Grazulis, Thomas P., Significant Tornadoes, 1880-1989. St. Johnsbury, VT: Environmental Films, 1991. Classic and fascinating two-volume reference detailing virtually every U.S. tornado F2 and greater for more than a century. Grazulis now runs The Tornado Project at http://www.tornadoproject.com/ .

HomeInsurance.com http://homeinsurance.com/home-insurance-101/what-is-personal-property-coverage.php estimates typical personal property as 50 to 70 percent the value of the dwelling structure itself. The New York State Department of Financial Services http://www.dfs.ny.gov/insurance/homeown/html/hmmaintn.htm says that 50 percent is often a basic percentage. Geico offers a free online Personal Property Cost Calculator http://www.geico.com/information/property-calculator/ .

Horton, A. H. and H. J. Jackson, The Ohio Valley Flood of March–April, 1913, Including Comparisons with Some Earlier Floods, (Department of the Interior, United States Geological Survey, Water-Supply Paper 334, Government Printing Office, Washington, D.C., 1913).

Lumsden, L.L., Sanitation of Flood-Stricken Towns and Cities, with Special Reference to Conditions Observed in River Towns and Cities of Kentucky, U.S. Public Health Service, Public Health Reports, Reprint No. 131, June 13, 1913; Washington, Government Printing Office, 1913.

McCampbell, E. F., “Special Report on the Flood of March, 1913,” reprinted from Monthly Bulletin Ohio State Board of Health, May 1913; pp 299–445.

National Flood Insurance Program fact sheet on flood insurance. http://www.floodsmart.gov/toolkits/spanish/downloads/english/facts-and-figures.pdf See also http://www.floodsmart.gov/floodsmart/pages/flooding_flood_risks/map_changes_flood.jsp for maps of flood risk.

Officer, Lawrence H. and Samuel H. Williamson, “Measuring Worth is a Complicated Question,” http://measuringworth.com/index.php ; for the actual calculator, see “Seven Ways to Compute the Relative Value of a U.S. Dollar Amount, 1774 to Present” http://measuringworth.com/calculators/uscompare/. See also their discussion “Choosing the Best Indicator to Measure Relative Worth” at http://www.measuringworth.com/indicator.php , using the cost of constructing the Empire State Building as an example.

O’Harrow, Dennis, “Indiana Flood Damage,” State Planning Board of Indiana, February, 1937. Typescript 2-page document on file with the Indiana State Archives, Indianapolis.

Railway Age Gazette, articles from March 28-April 25, 1913 on effects of flood on railroads, and recounts the estimates of the president, vice president, and chief engineer of the B&O as to its damages after their inspection around Ohio. Also consulted Bradstreet’s and Dun’s Review for April 1913.

Superstorm Sandy and Hurricane Katrina damages are compared at
http://www.cnn.com/2012/10/30/us/sandy-records. NOAA declares Katrina the costliest hurricane on record in "Hurricanes in History".

Bell, Trudy E., The Great Dayton Flood of 1913, Arcadia Publishing, 2008. Picture book of nearly 200 images of the flood in Dayton, rescue efforts, recovery, and the construction of the Miami Conservancy District dry dams for flood control. (Author’s shameless marketing plug: Copies are available directly from me for the cover price of $21.99 plus shipping, complete with inscription of your choice; for details, e-mail me at t.e.bell@ieee.org )