Introduction

My name is David Muhlhausen. I am Research Fellow in Empirical Policy Analysis in the Center for Data Analysis at The Heritage Foundation. I thank Chairman Benjamin L. Cardin, Ranking Member Lindsey Graham, and the rest of the committee for the opportunity to testify today about crimes against the homeless and the Hate Crimes Against the Homeless Statistics Act of 2009 (S. 1765). The views I express in this testimony are my own and should not be construed as representing any official position of The Heritage Foundation.[1]

My testimony focuses on the following points:[2]

• Policymakers should be cautious in accepting the validity statistics generated by homeless advocates;
• Crimes against the homeless have not risen to a level that requires formal data collection by the federal government;
• The homeless frequently commit violent and property crimes; and
• The Hate Crimes Against the Homeless Statistics Act of 2009 is unnecessary.

Policymakers should be cautious in accepting the validity statistics generated by homeless advocates . The determination of whether a social problem requires governmental action often hinges on measuring the social problem in question. Properly understanding a social problem requires accurately assessing its conditions and prevalence. For this reason, trustworthy statistics based on sound analysis are central to well-reasoned public policymaking.

Statistics are important in defining social phenomena as a problem and bringing the problem to the forefront of public debate. Thus, advocates of various causes are constantly attempting to generate startling statistics to grab the attention of the public and persuade policymakers to support their causes.

However, policymaking is too often influenced by shaky data and questionable statistics. A relevant example is the national estimate of the number of the homeless that affected the policy debate during the 1980s and 1990s. In 1982, Mary Ellen Hombs and Mitch Snyder of the Community for Creative Non-Violence (CCNV) estimated that 1 percent of Americans or 2.2 million Americans were homeless on any given night in 1980.[3] CCNV projected that the homeless could reach over 3 million individuals by 1983.[4] This 3 million homeless estimate was widely disseminated by the media, frequently without any objective scrutiny.[5] The methodology CCNV used to derive their estimates was based on information acquired from 100 homeless service organizations operating in 25 cities and states.[6] The CCNV methodology was not based on sound social science practices, yet it unnecessarily attracted public attention.[7]

According to Christopher Hewitt, Emeritus Professor of Sociology at the University of Maryland Baltimore County, “There is some information on the homelessness situation in several localities, but it is impossible to discover how the overall rate of 1% was calculated because in most places the rate is much lower. The 2.2 million figure appears to have been a ‘guesstimate’, pulled out of nowhere.”[8] The CCNV’s 2 to 3 million homeless figures were a striking contrast to the estimates derived by social scientists. The estimates by social scientists generally ranged from 300,000-400,000 homeless individuals on any given night during the mid-1980s and mid-1990s.[9]

Professor Hewitt was not the only social scientist to question the 2 to 3 million figures. The late Peter H. Rossi, Professor of Sociology Emeritus at the University of Massachusetts-Amherst, commented that “it is surely strange that guesstimates of the sort issued by local advocacy groups and the Community for Creative Non-Violence should be cited repeatedly in the press as serious and valid calculations.”[10] Under pressure, Mitch Snyder eventually admitted that the CCNV estimate was fabricated.[11]

My reason for beginning my testimony with this subject is to remind Congress that it needs to use credible numbers and analysis when making public policy decisions.

Crimes against the homeless have not risen to a level that requires formal data collection by the federal government. According to a National Coalition for the Homeless (NCH) report, Hate Crimes Against The Homeless: America’s Growing Tide of Violence (hereinafter referred to as the NCH report), asserts that “This years’ report has the horrifying distinction of being the deadliest in a decade, at forty-three reported homicides.”[12] While every case of a violent act committed against an innocent homeless person is tragic and should be prosecuted to the fullest extent of the law, the prevalence of these crimes do not rise to a level that requires formal data collection by federal, state, and local governments.

Policymakers should be skeptical of the conclusions presented in the NCH report. First, the NCH report uses a highly questionable methodology for estimating crimes against the homeless. Using a variety of sources, the cases of violence against the homeless identified in the NCH report appear to be primarily collected from media reports and homeless advocates.[13] Media coverage is not necessarily a good or accurate measure of any social problem.

Second, the NCH report fails to acknowledge the amount of crime committed by the homeless. The NCH report only focuses on crimes committed by “housed” or domiciled individuals against homeless individuals, while it excludes crimes committed by the homeless against other homeless individuals.[14] More importantly, the report ignores the amount of crime committed by the homeless against domiciled individuals.

Third, and most important, the analysis presented in the NCH report fails to make the case that the federal government should collect data on crimes against the homeless. Chart 1 presents the annual number of homeless murders counted in the NCH report from 1999 to 2009. The NCH report only counts murders of homeless individuals committed by domiciled persons. This leads one to naturally ask, “How many innocent domiciled and homeless individuals were murdered by the homeless?”

Over the course of 11 years, the NCH report counts 288 homeless murders with an average of less than 26.2 incidents per year. From 1999 to 2003 in Chart 1, the number of homeless murders counted by NCH fell from 49 to 8. From 2003 to 2009, the trend reversed. During this period, the number of homeless murders counted by NCH increased from 8 incidents to 43 incidents. While the highest number of homeless murders was recorded in 1999, NCH notes the number of homeless murders in 2009 is “[m]ost disturbing.”[15]

The presentation of the number of homeless deaths in the NCH report does not display the number of homeless murders relative to the total number of all murders recorded in the nation. Chart 2 presents the NCH homeless murder counts as a percentage of all murders recorded by the Federal Bureau of Investigation (FBI). In 2009, the FBI counted 15,241 murders in the United States.[16] When observing Chart 2, please notice that the y-axis (vertical axis) ranges from 0 percent to 10 percent. Due to homeless murders accounting for such a minuscule percentage of overall murders, setting the y-axis to the standard range of 0 percent to 100 percent would make the line representing the percentage of homeless homicides indistinguishable from the x-axis (horizontal axis).

As can be seen in Chart 2, NCH homeless homicides from 1999 to 2009 never accounted from more than 0.32 percent of total murders. In 2009, homeless murders were 0.28 percent of all murders. Conversely, all other murders accounted for 99.72 percent in 2009.

To better comprehend how tiny the number of homeless murders counted by NCH are compared to the total number of murders committed in the nation, Chart 3 individually represents the 15,241 murders recorded by the FBI in 2009. Each of the 15,241 individual squares represents a single murder that occurred in 2009. The darker-shaded squares in the lower right-hand corner represent the 43 homeless murders reported by NCH. While tragic, the minuscule number of homeless murders counted by NCH fails to rise to the level of a national problem that requires federal data collection. Homicides of young black males are a much more pressing problem facing our nation. In 2009, the FBI identified 639 black males under 18 years old were murdered in 2009.[17]

However, when comparing incidents of crime over time or across jurisdictions, the standard practice is to express incidents of crime as a rate. Thus, the NCH count of homeless murders is expressed as the rate of incidents per 100,000 residents in Table 1. The U.S. Department of Housing and Urban Development estimated that there were 643,067 homeless individuals during a single point-in-time in 2009.[18] The entire population of the United States was over 307 million residents in 2009.[19] Based on these population figures, the rate of homeless individuals murdered by domiciled individuals can be calculated.

The national murder rate of the homeless using the data provided by NCH translates into 6.7 incidents per 100,000 homeless persons in 2009. The national murder rate for the entire national population was 5.0 incidents per 100,000 residents. While the homeless murder rate is higher than the national rate, the difference is neither startling nor a justification for the federal government to begin formally collecting statistics on homeless murders.

Table 1: Homeless and National Murder Rate, 2009
	Homeless	Nation
Population	643,067	307,006,550
Murders	43	15,241
Murders Per 100,000	6.7	5.0
Sources: FBI, Uniform Crime Reports, 2009; National Coalition for the Homeless, Hate Crimes Against the Homeless: America's Growing Tide of Violence, August 2010; and U.S. Department of Housing and Urban Development, Office of Community Planning and Development, The 2009 Annual Homeless Report, (Washington, D.C., June 2010).

After reviewing the national trend in homeless murders, NCH presents a state-by-state breakdown of violent crime committed by domiciled individuals against the homeless. According to NCH, there was at least one violent act committed against the homeless in 23 states and the District of Columbia during 2009. For the remaining 27 states, the NCH failed to record any violent acts against the homeless.

Table 2: Rate of Violent Acts Committed Against the Homeless by State, 2009
	State		Violent Acts Against the Homeless	Total Violent Crimes	Homeless Violent Act Rate (Per 100,000 Homeless)	Total Violent Crime Rate (Per 100,000 Residents)	State Homeless Population	Total State Population
1	Alaska	6		4,421	301.20	632.95	1,992	698,473
2	South Carolina	5		30,596	111.78	5,102.25	4,473	599,657
3	Ohio	13		38,332	102.36	1,001.97	12,700	3,825,657
4	Arkansas	2		14,959	70.13	565.97	2,852	2,643,085
5	New Hampshire	1		2,114	60.79	18.31	1,645	11,542,645
6	Utah	2		5,924	52.70	129.88	3,795	4,561,242
7	Oregon	8		9,744	46.22	52.56	17,309	18,537,969
8	Oklahoma	2		18,474	41.34	1,394.71	4,838	1,324,575
9	Mississippi	1		8,304	35.75	22.47	2,797	36,961,664
10	Dist. of Columbia	2		8,071	32.11	289.85	6,228	2,784,572
11	Florida	16		113,541	28.78	3,929.50	55,599	2,889,450
12	Illinois	4		64,185	28.46	1,740.82	14,055	3,687,050
13	Pennsylvania	4		47,965	26.50	954.58	15,096	5,024,748
14	Nevada	3		18,559	20.72	628.69	14,478	2,951,996
15	California	27		174,459	20.28	2,770.84	133,129	6,296,254
16	Texas	7		121,668	19.04	965.25	36,761	12,604,767
17	Tennessee	2		42,041	18.99	325.64	10,532	12,910,409
18	New Jersey	2		27,121	15.19	406.97	13,169	6,664,195
19	Missouri	1		29,444	14.37	118.81	6,959	24,782,302
20	Indiana	1		21,404	14.32	245.80	6,984	8,707,739
21	Colorado	2		16,976	13.10	283.52	15,268	5,987,580
22	Washington	2		22,056	8.78	343.38	22,782	6,423,113
23	Georgia	1		41,880	4.91	426.08	20,360	9,829,211
24	New York	1		75,176	1.64	384.70	61,067	19,541,453
Sources: National Coalition for the Homeless, Hate Crimes Against the Homeless: America's Growing Tide of Violence, August 2010; FBI, Uniform Crime Reports, 2009; and U.S. Department of Housing and Urban Development, Office of Community Planning and Development, The 2009 Annual Homeless Report (June 2010).

When looking at the states, the NCH report concludes that “Florida and California are the states where violence against the homeless occurs most frequently.”[20] NCH “speculates” that “the homeless population in these warmer areas is an easy target for hate crime aggressors. Generally, our data suggests that warmer year-round climates with higher concentrations of homeless individuals document more violent acts against the homeless.”[21] However, NCH’s analysis does not control for the size of the homeless populations within the states. Controlling for the size of the homeless populations within the 23 states and the District of Columbia allows us to determine which jurisdictions have higher rates of violent acts against the homeless.

Table 2 uses NCH and FBI data to present the violent crime rate against the homeless and general populations of the 23 states and the District of Columbia. The states are ranked by highest rate of violent acts against the homeless. While the NCH report singled out California and Florida for having the highest number of violent acts against the homeless, these states rank in the middle compared to the other states when the homeless population is taken into account. Ranked 15^thout of 24 jurisdictions, California had 20.28 violent acts against the homeless per 100,000 homeless individuals. With a violent act rate of 28.78, Florida ranked 11^th among the jurisdictions. The state with the highest rate of violent acts against the homeless is Alaska. Controlling for the homeless population certainly undermines NCH’s hypothesis that warmer climates make the homeless more vulnerable to violent crimes.

Table 2 also presents the violent crime rate for each of the states and the District of Columbia. In general, the violent crime rates are higher than the rates of violent acts against the homeless. The exception is New Hampshire and Mississippi. Each of these states had a higher rate of violent acts against the homeless than their overall violent crime rate.

When homeless advocates claimed that there were 3 million homeless individuals in America at any given point-in-time during the early 1980s, the number was so startling that it caught the attention of policymakers and the media. While the NCH report interpreted its data as presenting “shocking,” “alarming,” and “disturbing” findings, it produces no such startling numbers to support its goal of persuading federal, state, and local governments to collect data on crimes against the homeless.[22] While crimes against the homeless are unfortunate and should be prosecuted to the fullest extent of the law by state and local governments, the data presented in the report does not match the level of rhetoric used in the report.

The homeless commit too many violent and property crimes. While the homeless are frequently victims of crime, the NCH report conveniently failed to address the prevalence of crime committed by the homeless. According to social science research, the homeless are generally not a collection of law abiding individuals.[23]

A survey of 432 homeless youth between the ages of 13 and 23 years of age living in the Hollywood area of Los Angeles found that 25 percent admitted to having attacked another person with a knife (17 percent since residing on the streets) and 22 percent reported having fired a gun at someone (14 percent since residing on the streets).[24] Another study of 200 homeless youth residing in Edmonton, Alberta, Canada, found that they reported, on average, committing 3.11 property crimes, 2.89 drug deals, and 0.87 robberies over an undisclosed period of time.[25]

Of 42 homeless youth in San Jose, California, 96 percent reported contact with the police (40 percent reported frequent police contact), 74 percent had been previously incarcerated in juvenile correctional facilities, 31 percent were gang affiliated, and 73 percent reported having stolen from a store or person.[26]

A random sample of arrest data from Austin, Texas, found that homeless males accounted for:

• 4.0 percent of violent arrests;
• 0.0 percent of murder arrests;
• 6.5 percent of rape arrests;
• 9.6 percent of robbery arrests;
• 1.4 percent of aggravated assaults;
• 9.5 percent of property crime arrests;
• 12.1 percent of burglary arrests
• 8.7 percent of larceny-theft arrests; and
• 9.5 percent of auto theft arrests.[27]

The same study compared the arrest rates of homeless men to the arrest rates of domiciled men. While the study found that domiciled men had higher rates of arrest for overall violent crime and aggravated assault, homeless men had statistically higher arrest rates for robbery.[28] The difference in murder and rape arrests rates between homeless males and domiciled males were statistically indistinguishable.[29] For property crimes, homeless men had statistically higher rates of arrest for overall property offenses, burglaries, larceny-thefts, and auto thefts.[30]

The Hate Crimes Against the Homeless Statistics Act of 2009 is unnecessary. When Congress considers the need for collecting data on any social phenomena, the nature of the evidence presented to Congress should be instrumental to the decision-making process. A wrong assessment of the evidence can lead Congress to waste valuable resources of the federal, state, and local governments. The NCH draws conclusions far beyond the data presented in its report. An objective and fair analysis of the data presented in the NCH report simply does not support the notion that the federal government needs to collect statistics on crimes committed against the homeless by domiciled individuals.

The amount of crimes committed against the homeless by domiciled individuals, let alone such incidents motivated by “hate,” does not rise to the level of requiring the federal government to collect statistics on this issue. Crimes against the homeless, like all other ordinary street crimes, should be prosecuted to the fullest extent of the law by state and local governments.

While some may argue that the lack the of reliable data on the number of crimes committed against the homeless by domiciled individuals is justification enough for federal intervention, such logic leads the federal, state, and local governments down the road of collecting data on any perceived social problem, whether the problem warrants attention or not. The Hate Crimes Against the Homeless Statistics Act of 2009 is unnecessary.

Endnotes

[1]Although all opinions expressed and any errors herein are my own, I am thankful for Senior Data Graphics Editor John Fleming’s assistance with the graphics for this testimony.

[2]While my testimony concentrates on the statistics used to advocate in support of the Hate Crimes Against the Homeless Statistics Act of 2009 (S. 1765), it should be noted that the congressional regulation of hate crimes, whether against homeless individuals or others raises serious constitutional questions arising from the lack of a plenary police power in Congress. See, e.g., Brian Walsh, “Federal Hate Crimes Statute: An Unconstitutional Exercise of Legislative Power,” Heritage Foundation WebMemo #2416, April 29, 2009, at http://www.heritage.org/Research/Reports/2009/04/Federal-Hate-Crimes-Statute-An-Unconstitutional-Exercise-of-Legislative-Power.

[3]Mary Ellen Hombs and Mitch Snyder, Homeless in America: A Forced March to Nowhere (Washington, D.C.: Community for Creative Non-Violence, 1982).

[4]Ibid.

[5]Christopher Hewitt, “Estimating the Number of Homeless: Media Misrepresentation of an Urban Problem,” Journal of Urban Affairs, Vol. 18, No. 3(1996), pp. 431-447.

[6]Ibid., p. 432.

[7]Ibid.

[8]Ibid.

[9]Ibid., p. 434.

[10]Peter H. Rossi, Down and Out in America: The Origins of Homelessness (Chicago: University of Chicago Press, 1989), p. 70.

[11]Gary S. Becker, “How the Homeless ‘Crisis’ was Hyped,” in The Economics of Life, eds. Gary S. Becker and Guity Nashat Becker (New York: MacGraw-Hill, 1997), pp. 175-176.

[12]National Coalition for the Homeless, Hate Crimes Against the Homeless: America’s Growing Tide of Violence, August 2010, p. 8, at http://www.nationalhomeless.org/publications/hatecrimes/hatecrimes2009.pdf.

[13]Ibid., p. 10.

[14]Ibid.

[15]Ibid., p. 11.

[16]Federal Bureau of Investigation, Uniform Crime Reports, 2009, at http://www.fbi.gov/ucr/cius2009/offenses/violent_crime/murder_homicide.html (September 26, 2010).

[17]Federal Bureau of Investigation, Uniform Crime Reports, 2009, Expanded Homicide Data Table 2, at http://www.fbi.gov/ucr/cius2009/offenses/expanded_information/data/shrtable_02.html (September 26, 2010).

[18]U.S. Department of Housing and Urban Development, Office of Community Planning and Development, The 2009 Annual Homeless Report (June 2010), p. i.

[19]Federal Bureau of Investigation, Uniform Crime Reports, 2009.

[20]National Coalition for the Homeless, Hate Crimes Against the Homeless, p. 14.

[21]Ibid., p. 14.

[22]Ibid., pp. 8 and 11.

[23]Michele D. Kipke, Thomas R. Simon, Susanne B. Montgomery, Jennifer B. Unger, and Ellen F. Iversen, “Homeless Youth and Their Exposure to and Involvement in Violence While Living on the Streets,” Journal of Adolescent Health, Vol. 20 (May 1997), pp. 360-367; Stephen W. Baron, “Street Youths and Substance Abuse: The Role of Background, Street Lifestyle, and Economic Factors,” Youth & Society, Vol. 31, No. 3 (1999), pp. 3-26; Marci Schwartz, Heather Kissinger Sorensen, Seth Ammerman, and Emilie Bard, “Exploring the Relationship Between Homelessness and Delinquency: A Snapshot of a Group of Homeless Youth in San Jose, California,” Child and Adolescent Social Work Journal, Vol. 25 (2008), pp. 255-269; and David A. Snow, Susan G. Baker, Leon Anderson, “Criminality and Homeless Men: An Empirical Assessment,” Social Problems, Vol. 36, No. 5 (1989), pp. 532-549.

[24]Kipke et al., “Homeless Youth and Their Exposure to and Involvement in Violence While Living on the Streets,” p. 363.

[25]Baron, “Street Youths and Substance Abuse,” p. 13.

[26]Schwartz et al., “Exploring the Relationship Between Homelessness and Delinquency,” p. 263.

[27]Snow et al., “Criminality and Homeless Men: An Empirical Assessment,” p. 537.

[28]Ibid., p. 538.

[29]Ibid.

[30]Ibid.

***************************************************************************

The Heritage Foundation is a public policy, research, and educational organization recognized as exempt under section 501(c)(3) of the Internal Revenue Code. It is privately supported and receives no funds from any government at any level, nor does it perform any government or other contract work.

The Heritage Foundation is the most broadly supported think tank in the United States. During 2013, it had nearly 600,000 individual, foundation, and corporate supporters representing every state in the U.S. Its 2013 income came from the following sources:

Individuals 80%

Foundations 17%

Corporations 3%

The top five corporate givers provided The Heritage Foundation with 2% of its 2013 income. The Heritage Foundation’s books are audited annually by the national accounting firm of McGladrey, LLP.

Members of The Heritage Foundation staff testify as individuals discussing their own independent research. The views expressed are their own and do not reflect an institutional position for The Heritage Foundation or its board of trustees.