Methodology in Housing Policy

Despite the worsening of the housing crisis in North America — the Canadian and US federal governments refuse to discuss the issue in any meaningful way. Still, this crisis deserves special attention, so what have authors said about the matter? Constantine Kontokosta and Alan Walks and Brian Clifford provide compelling arguments on housing policy and political economics. The former employs a comparative analysis of regional and local governments in America, and the latter uses a descriptive case study to ascertain Canada’s federal management of the housing market. Furthermore, we argue that no single research method suffices; instead, political scientists should balance their approaches to create a comprehensive methodology. Such a methodology provides insightful findings while ensuring an argument’s intelligibility and measurement validity. This thesis may be overdone, but its very truth and importance call for an inquiry. As severe as the housing crisis is, political scientists expect a well-rounded methodology. Hence, to test this argument, Kontokosta and Walks and Clifford make for a good comparison. To this end, we summarize and compare the above two articles. They both use case studies; however, Kontokosta couples his with statistical analysis and Walks and Clifford develop their argument with document analysis: the former’s case study is selective, the latter’s is comprehensive. Both studies’ methodological strength lies in their insights, unearthing successes and failures in the recent history of housing policy. Citation: Brooks C, Praud J (2018) Methodology in Housing Policy. Arts Social Sci J 9: 342. doi: 10.4172/2151-6200.1000342


Introduction
Despite the worsening of the housing crisis in North America the Canadian and US federal governments refuse to discuss the issue in any meaningful way [1,2]. Still, this crisis deserves special attention, so what have authors said about the matter? Kontokosta [3], Walks and Brian [4] provide compelling arguments on housing policy and political economics 1 . The former employs a comparative analysis of regional and local governments in America, and the latter uses a descriptive case study to ascertain Canada's federal management of the housing market 2 .
Furthermore, we argue that no single research method suffices; instead, political scientists should balance their approaches to create a comprehensive methodology. Such a methodology provides insightful findings while ensuring an argument's intelligibility and measurement validity. This thesis may be overdone, but its very truth and importance call for an inquiry. As severe as the housing crisis is, political scientists expect a well-rounded methodology. Hence, to test this argument, Kontokosta [3], Walks and Brian [4] make for a good comparison.
To this end, we summarize and compare the above two articles. They both use case studies; however, Kontokosta [3] couples his with statistical analysis (2015, 575-576) and Walks and Brian develop their argument with document analysis (2015,(1640)(1641)(1642): the former's case study is selective, the latter's is comprehensive. Both studies' methodological strength lies in their insights, unearthing successes and failures in the recent history of housing policy (2015, 587; 2015, 1624). Second, we critique these articles with the broader literature. With authors like Hansson [5] drawing on Kontokosta [3], Wetzstein 1 Kontokosta's is "Do inclusionary zoning policies equitably disperse affordable housing? A comparative spatial analysis." Walks and Clifford's is "The political economy of mortgage securitization and the neoliberalization of housing policy in Canada." 2 Robert Yin defines a case study as "an empirical enquiry that investigates a contemporary phenomenon within its real-life context, especially when the boundaries between phenomenon and context are not clearly evident" [13]. As Yin argues, "the case study allows an investigation to retain the holistic and meaningful characteristics of real-life events" [13].
[6] upon Walks and Brian, their case studies provide utility to housing policy scholarship, Third, we discuss their methodological choices. For example, both articles use process tracing, and so their internal validity is high, but without sufficient conceptualization of their core concepts and measures, their construct validity is doubtful and their arguments are unintelligible 3 . Interviews would also assist in this conceptualization to validate their constructs [7,8]. Lastly, while they cannot detract from their case studies, they need to differentiate they terminology, use familiar language, and reduce the volume of jargon [9]. We finish with recommendations about how these additions might occur, with reference to both articles' research objectives.

Summary and Comparison of Articles
Kontokosta employs a comparative case study to analyze the landuse policies of regional and local governments in the US. Throughout the 1980s and 1990s, American officials faced public scrutiny over racial segregation and socio-economic status, as land-use was a contentious issue [3] 4 . Regional or local politicians in the US often faced pressures from lower socio-economic classes and minority racial groups about affordable housing and poverty [3]. Officials wanted to disperse poverty in lower class regions by developing affordable housing in higher-class areas [3]. These officials required a housing tool. Hence, they implemented one that forces developers to build affordable housing in the middle-to high-class areas [5]. This tool is the inclusionary zoning (IZ) policy.
The IZ policy is the core concept in Kontokosta [3]. This policy's purpose is to "disperse affordable housing throughout a region or municipality by leveraging market-rate development to subsidize new construction" [3] 5 . He argues, "IZ policies emerged in the 1970s as a means for local governments to meet their responsibility for affordable housing provision while fostering neighbourhood integration". US officials implemented IZ policies as a part of their mixed-income affordable housing strategy, with the impression that socio-economic integration results in both intended and unforeseen positive outcomes.
Nevertheless, a rigorous evaluation of IZ policies is missing from the affordable housing literature. This gap is Kontokosta's research problem. Hence, he assesses how well these policies have equitably distributed "low-income housing geographically" by comparing two regions in the US: Montgomery County, Maryland, and Suffolk County, New York (2015, 570). Montgomery is a regional jurisdiction, and Suffolk is a local jurisdiction. He measures the effectiveness of IZ policies by calculating how many low-income units these counties built between 1980 and 2000 [3]. Lastly, he judges IZ policies as failures if they only increased the density of poor people in low-income areas.
Furthermore, his construct is IZ units, and his construct measure is the number of IZ units built in Suffolk and Montgomery Counties between 1980 and 2000. Since he cannot analyze all IZ units constructed in these counties for this time, he examines more than IZ 11,000 units in both counties to determine statistical differences in IZ unit construction (2015, 569). To make this calculation, he generates official statistics. He develops a "geo-coded property-specific database of more than 11,000 IZ units" (2015, 569) 6 . With geo-coding, he establishes the first census tracts for IZ unit production and spatial clustering (2015, 569) 7 . He recognizes this data generation as a strength to his research. Before these statistics, there was no empirical evidence of the impact of IZ policies, and without these figures, officials cannot evaluate past implementations of IZ policies [3]. To create this tract, he also gathers statistics from multiple sources, such as "Suffolk Department of Planning" and "Housing Opportunities Commission" (2015, 576). While Kontokosta notes that the results for his analysis are not necessarily generalizable for larger populations, his aim was not to provide a comprehensive analysis but to inform policy-makers about the potential effectiveness and limitations of IZ policies.
Comparative case studies often test theories or conjectures, and so Kontokosta uses this data and an analysis of relevant literature to test four propositions 8 . First, IZ programs implemented by regional governments coordinated with land-use policies and "overcame local opposition" to the siting of IZ units [3] 9 . Second, the regional government's IZ programs had a stable program design and avoided program variation, stability that ensured the spatial distribution of IZ units across differing socio-economic areas [3] 10 . Third, developers built IZ units in proportion to the number of residences in an area and even a higher percentage of neighbourhoods with more multi-family 7 Census tracts are official demographic statistics for relatively stable but small geographic areas [26]. 8 See Kontokosta (2015, 572-574) for the full description of these propositions. 9 Programs develop from policies. 10 Local governments may vary the IZ programs.
housing [3]. Forth, officials were more likely to coordinate regional IZ programs with regional Smart Growth policies than local IZ programs [3] 11 . IZ policy impact depends on program structure, the political will to execute IZ policies, the presence of complementary land use policies, the progress of the local housing market, and the degree of community opposition to development [3]. He tests these propositions and then discusses their policy implications.
After testing these propositions, there is one major finding: IZ policies are disproportionately more successful when conducted by the regional government in Montgomery County than the local government in Suffolk County. He finds that there was seventy-five percent more low-income housing built in Montgomery County than Suffolk County within his census tract [3]. Suffolk County was without regional government oversight and was influenced by antihousing development locals from high socio-economic classes, whereas Montgomery County maintained regional authority over housing policy and was not under this influence, and so its IZ policies were implemented effectively [3]. Though there are other relevancies, regional government oversight is the most important.
Similar to Kontokosta, Walks and Brian employ a case study, but theirs is a descriptive case study on Canada's economic policies for the housing market between 1985 and 2011 12 . The crisis started with the Canadian government's desire to off-load the construction of affordable housing to the private sector [4]. The government took a minimalist approach, merely regulating the housing market instead of building social housing [4]. According to Walks and Brian, scholars knew little about these regulations, despite their catastrophic impact. Hence, Walks and Brian wanted to provide some elaboration.
Their research question is, how can political economists understand the processes of financialization and neoliberalization in Canada's housing policy? To answer this question, Walks and Brian focus on mortgage securitization, a financial control that shifts mortgage risk from the lender to another person or group, so that lenders assume little to no liability [4]. This control encourages private persons to fund loans, and thereby "facilitates increased access to credit" [4]. Mortgage securitization is the central concept of Walks and Brian's case study.
This policy transferred the risk from lenders to the Canadian public. When someone defaults, the Canadian government must reimburse lenders ninety to one hundred percent of the remaining debt [4]. From 1985 to 2011, the government chose this policy to encourage lenders to finance potential homebuyers who could not afford a house otherwise [4]. These lenders financed these homes for exceptionally high prices, which drove up household debt and the cost of homes in Canada [4]. Easy access to credit increased housing demand, and this demand inflated housing prices.
The history of mortgage securitization in Canada is the context of Walks and Brian's case study, and to show this history, Walks and Brian analyze secondary literature and government documents. Government reports supply data to explain the rise in outstanding credit that the Canadian government owned from 1987 to 2011 [4]. Analyzing this documentation also uncovers the "political and legislative events through which Canada's peculiar state-driven draws critical attention? Should these articles complement expository modes with their case studies? To answer this question, scholars must understand the strengths and limitations of Walks and Brian [4] and Kontokosta [3]. Both articles' methods contribute to the housing policy and political economics literature. Walks and Brian's evidence helps substantiate current literature in the field. In citing Walks and Brian (2015), Wetzstein argues that the Canadian government regulates the housing market to avoid social housing commitments (2017, 3166). Instead of developing housing infrastructure, the government merely regulates the private sector's housing initiatives. Wetzstein ties this neoliberal approach to housing with the global austerity policies that followed the financial crisis (2017, 3166). Worldwide people face eviction due to welfare cutbacks (2017, 3160). There are fewer jobs because austerity fails to create them. Canada is no exception: it also enacted austerity policies, coupling its cutbacks in welfare spending with minimal involvement in the housing market [10]. With Walks and Brian's policy evaluation, Wetzstein provides a commentary on austerity policies.
Besides excavating the evidence of Canada's failures in housing policy, Walks and Brian advance a research agenda. In Western academia, scholars usually see housing as a city-planning concern. Walks and Brian [4] set a precedent: they tie the study of housing policy to political economics. Following Walks and Brian's example, Wetzstein exhorts: we must "bring back political economy" into questions of housing policy (2017, 3163). He continues, if scholars wish to understand housing policy, they must analyze the strategic processes both of global markets and state-regulatory bodies (2017, 3163). Walks and Brian's evidence strengthens Wetzstein's argument.
Walks and Brian are not alone in their contribution to the broader literature, Kontokosta's research is also forceful. Drawing on Kontokosta's argument, Hansson points out that since IZ policies are so uncertain, to combat housing crises, "public subsidies are also necessary" (2017, 472). She argues that governments need to subsidize the construction industry to ensure housing development. To ground this position, Hansson invokes Kontokosta's argument: "the effectiveness of such policies is dependent on program structure, political will to enforce policy,… and potential opposition to development" [5]. Kontokosta reveals the limitations of IZ policies, and Hansson accounts for these weaknesses to determine what other policy interventions are necessary to ensure positive outcomes with IZ policies. Ultimately, how scholars like Hansson apply Kontokosta's argument reinforces the strength of his case study.
Nevertheless, one shortcoming of Kontokosta's case study is its timeframe 15 . Though Kontokosta's comparative case study focuses on the timeframe 1980-2000, reference to the broader context of housing crises in North America is wanting. we agree with Kontokosta's selective timeline to evaluate IZ policies, but he could have related the 2008 Financial Crisis to his narrative [3]. In "The housing bubble and the financial crisis," Dean Baker argues that the 2008 Financial Crisis not only ruined the US housing market, but it originated in the 1990s, which falls within Kontokosta's case study timeframe (2008, 73). The financial crisis is a necessary background factor because it devastated the US housing market.
Kontokosta's discussion need not digress. Merely introducing this crisis would have contextualized the rise in popularity of IZ policies 15 The difference between case studies and history is that the former focuses on contemporary events, whereas the latter has no timeframe restrictions in studying the past [13]. Nevertheless, case study approaches overlap with historical methods, and sometimes the broader history can contextualize a case study. program of securitization was fashioned" [4]. The government pursued mortgage securitization with a trial-and-error effort [4]. From 1992 to 2001, the Canadian government took up aggressive deficit reduction amid the 1990s recession and capped social housing subsidies. These austerity measures contributed to the 2008 Financial Crisis, and then the government experimented with mortgage securitization to try to solve the crisis [4]. Though the government introduced mortgage securitization policies in the 1990s, it continued to experiment with them after the financial crisis. With government documentation, Walks and Brian unearth the chronology of securitization program failures, a timeline that reveals the housing crises that neoliberalization creates. Both Walks and Brian and Kontokosta evaluate housing policies, though for slightly different reasons. Since there is a discrepancy about the causation of the housing crisis, Walks and Brian investigate Canada's neoliberal housing policies before and after the 2008 financial crisis (2015, 1639-1640). They wish to inform scholars, housing policymakers, and analysts about the fallout of these policies, as they were haphazard and experimental. In contrast, Kontokosta generates evidence of past affordable housing policies, so that policy-makers can appreciate the strengths and weaknesses of IZ policies (2015, 569-570). Kontokosta hopes that policy-makers will consider this evaluation, and possibly supplement IZ policies with other interventions to secure positive outcomes when IZ policies need extra support (2015, 569, 587-588). Despite these differences, both articles evaluate housing policies.
Process tracing is fundamental to both these cases, as they investigate the underlying causation of past housing policies 13 . Kontokosta analyzes IZ policy operations with official documents and census data (2015, 569-570). Walks and Brian trace the processes of the Canadian housing crisis using document analysis, a financial policy timeline, and secondary literature (2015, 1629) 14 . Thus, these articles connect several policies, programs, and events to establish a causal chain. Hence, both studies have substantial evidence to support their arguments.
Nevertheless, these articles' results sharply differ. Kontokosta develops new census tract data through geo-coding of more than 11,000 IZ units [3,9]. These census tracts supply policy-makers with new empirical evidence concerning the number of IZ housing units built from 1980 to 2000. Using geo-coding, he found that Montgomery developed more IZ units than Suffolk County in this census tract from 1980 to 2000 (2015, 586). In contrast, Walks and Brian do not create primary data, but analyze official documents and secondary data, mostly from the broader political economics and neoliberal literature (2015,(1640)(1641)(1642). This data establishes the timeline and narrative for their case study, providing official dates and corresponding expenditures of neoliberal housing programs and policies (2015, 1632, 1635, 1637). This method leads to the finding that the housing crisis in the 2000s derived from neoliberal policy experimentations in the 1980s. Kontokosta bases his evaluation on quantitative results, whereas Walks and Brian found theirs on interpretations of mortgage securitization history.

Critical Analysis of Articles
What is it about Walks and Brian's and Kontokosta's methods that 13 Process tracing incorporates a multitude of sources to verify a cause's processes [9], which coordinates with case study inquiry, for case studies rely "on multiple sources of evidence, with data needing to converge" [13]. 14 Glenn Bowen defines document analysis as a "systematic procedure for reviewing or evaluating documents -both printed and electronic (computer-based and Internet-transmitted) material. Like other analytical methods in qualitative research, document analysis requires that data be examined and interpreted in order to elicit meaning, gain understanding, and develop empirical knowledge" [14]. and the need to evaluate them [5]. The political scientist will ask: since the 2008 Financial Crisis destabilized the US housing market, are IZ policies sufficient to meet affordable housing demands [5]? Answering this question would reinforce the need to evaluate IZ policies, for their success depends on local US housing markets, which the financial crisis destabilized [3,11]. Even then, Kontokosta omits reference to the financial crisis.
Kontokosta could have also discussed the failures of neoliberal housing policies to justify his evaluation. Hansson argues that IZ policies became popular in the 1990s because "housing policy in much of the Western world became a local rather than national policy issue, and there was a general trend toward more market-based solutions" (2017, 472). Nonetheless, these market-based solutions were not only IZ policies but also the neoliberal housing policies that sparked the 2008 Financial Crisis [4,10] 16 . Kontokosta is missing some vital background information to understand the need for IZ policies. Without relevance to neoliberal policy failures and the financial crisis, the significance of IZ policies is difficult to assess.
A brief analysis of Kontokosta's [3] and Walks and Brian's [4] scholarly context demonstrates their methodological utility's extent, but to add to the critique, assessing both articles in their own right will prove good. Both Kontokosta and Walks and Brian deal with a small number of cases. Their internal validity is high because they do not conflate the processes they analyze but explicate them [7]. Kontokosta focuses on Montgomery and Suffolk County. He considers socioeconomic status, regional and local authorities, political culture, and constructed IZ units [3]. Walks  Furthermore, Walks and Brian are comprehensive with their document analysis, whereas Kontokosta is selective with his generation of census data. Comparative case studies aim not to generalize results from a sample to a population, but to investigate the causation and context of events [12,13] 17 . Kontokosta clarifies that his statistical analysis is not for generalization but to show that similar IZ policies in Montgomery County and Suffolk County had "significant variation" in outcomes (2015, 588). He compares these policies not to prove perfect generalization for similar cases, but to inform officials and scholars of these policies' impact in Montgomery and Suffolk County, information that suggests how effective these policies would be under similar circumstances.
On the other hand, Walks and Brian's study is comprehensive. Walks and Brian created a chronology from nineteen government documents, including seven reports, standing committee proceedings, legislative summaries, program evaluations, and housing price indices. Though this timeline was quantitative, these records also provided materials for a qualitative analytical commentary [4]. With these documents, Walks and Brian "elicit meaning, gain understanding, and develop empirical knowledge" [14]. Lastly, Walks and Brian reinforce this document analysis with secondary data and background 16 Even Kontokosta notes, "the original intent of IZ policies was to … promote affordable housing with limited subsidy" [3]. 17 Though comparative case studies can generalize theories or propositions to other contexts [27].
information from journals, books, and news media. This extensive research makes their analysis comprehensive.
Further, the case study is choice worthy for both articles because it produces insightful findings in analyzing causation and process. Yin agrees, "case studies are the preferred strategy when 'how' or 'why' questions are being posed" (2003, 1). How questions concern process analysis; and, the case study ascertains a process' "causal mechanisms" [15]. The case study answers why questions because it investigates causation, verifying or disproving why an event occurred, as opposed to detailing a series of steps [16]. Though causation underlies processes, processes concern a chain of causes instead of concentrating intensely on the reasons behind a single cause. Process and causation tend to coordinate, and the case study is open to both.
In Walks and Brian [4], their case study fits because their overarching topic is how and why the housing crisis occurred. The outbreak of the housing crisis was obscure due to misperceptions about the effectiveness of neoliberal housing policies [4]. These authors show the Canadian banks' recklessness in experimenting with mortgage securitization policies. They also reveal corruption in Canadian officials, who portrayed these procedures as successes when they were failures (2015, 1634, 1638, 1639). The case study explains how and why the housing crisis broke out, and the findings bring these neoliberal policies into disrepute.
Further, the case study allows Walks and Brian to evaluate these policies on a timeline, which reveals their trial-and-error history. They find that "[t]he confluence of events encouraged trial-and-error policy experimentation" (2015, 1639). The confluence of events began with the onset of neoliberal austerity and the cancellation of social housing programs from 1984-1992, an austerity measure that triggered the housing crisis and forced policy-makers down a trial-and-error path [4]. Walks and Brian's case study uncovers these processes and is therefore defensible.
The comparative case study is choice worthy when researchers need to distinguish among competing causal explanations. Explanations can be ambiguous [15]. Take the example of Kontokosta. He sets out to evaluate IZ policies but assessing the extent to which intervening "factors" influence the impact of individual policies is difficult (2015, 570) 18 . He is right to choose the comparative case study because he needs to "explain how features within the context influence the success of programmes or policy initiatives" [12]. Kontokosta's analysis of official statistics for Suffolk County, from 1980 to 2000, show that developers concentrated nearly all IZ units in low-income neighbourhoods (2015, 587). Not only did developers in Montgomery County build seventyfive percent more IZ units than in Suffolk County from 1980 to 2000 in his census tract, but they also dispersed them across middle-to highclass areas [3]. Suffolk County designed their IZ programs without regional oversight, which suggests their IZ policies failed because of its local government. Kontokosta's comparative case study puts weaker causal explanations aside and focuses on this finding, a finding only possible from this comparison. they measure [17]. Thus, their construct validity is uncertain [18]. This insufficiency compromises their measurement validity and puts into question whether their methods can measure their concepts and variables 19 .
In both Kontokosta [3] and, Walks and Brian [4], their construct validity is uncertain because they fail to conceptualize their construct measures. Kontokosta's research objective is to assess the impact of IZ policies in Suffolk and Montgomery Counties. To evaluate this impact, he needs to understand why policy implementation differed in these counties. He argues that this impact depends on a series of factors including "the political will to enforce" IZ policies and the "strength of the local housing market," but he omits explaining or defining these construct measures, which compromises his construct validity (2015, 572). For instance, political will is ambiguous; it could mean administrative commitment or ambition to execute policies. Nevertheless, whether he intends commitment or ambition, it is uncertain how he assesses policy-makers with these measures, and, therefore, their IZ policy implementation (2015, 572-574). His central construct is IZ units, and his measure for this construct is the number of IZ units built in Suffolk and Montgomery Counties from 1980 to 2000 (2015, 569). He can determine statistical differences in IZ units built, but he cannot comprehend implementation differences in these counties because definitions of political will and housing market strength are absent [18]. To increase his construct validity and measure differences in policy implementation, he needs to conceptualize these construct measures.
Kontokosta's construct measures are questionable but just as problematic is Walks and Brian's. Their research objective is to explain neoliberalization's ineffectiveness by showing its parallels with failures in securitization policies (2015, 1624-1625). To illustrate this ineffectiveness, they argue that Canadian policy-makers pursued securitization policies in a path-dependent, trial-and-error manner (1625). Their construct is securitization, and their construct measure is path-dependence. While Walks and Brian define securitization, they do not explain or define path-dependence 20 . They suggest path-dependence occurs when officials lack control over the intention of their policies since political structures, "not of their own making," influence policyformation (2015, 1625). For example, he argues that policy-makers in the 1980s were locked-in to securitization policies due to the Canadian government's bias towards social housing policies (2015, 1631). The government blamed the 1980s recession, in part, on these social policies, a structural force that compelled policy-makers to experiment with securitization policies [4]. Nevertheless, this path-dependent policy description is entirely inferential, as he provides neither conceptualization nor explanation. He argues that neoliberalization is ineffective since its path-dependent, securitization policies were experimental and resulted in Canada's housing crisis, but how he measures securitization is unclear because his construct measure, pathdependence, needs conceptualization.
Conceptualizing path-dependence is a start, but to further enhance its meaning and parameters Walks and Brian could have interviewed significant monetary officials. An interview with both qualitative and quantitative questions about path-dependence, whether formal or flexible, may offer inside information to conceptualize it [19]. Walks and Brian merely allude to policy actors in their case study, when interviewing them with questions about path-dependence over the last thirty years gives the term context and meaning. For instance, they report that the Canadian banking deputy governor characterized the housing market as unstable due to the path-dependence of the Canadian Mortgage and Housing Corporation (2015, 1637). Possibly, they could strengthen their case study by interviewing this actor, who might explain and interpret path-dependence. Whether Walks and Brian had the means to interview this official is another matter [19], but integrating interviews into their case study would support a conceptualization of path-dependence, for its meaning would derive from relevant policy actors. With this information, path-dependence might then become a valid measure, which could make securitization measurable.
Additionally, the interview would prove useful in Kontokosta's case study. In addition to conceptualizing political will to enforce IZ policies, Kontokosta could have interviewed relevant public officials to understand the extent of their political will to implement IZ policies. Understanding extent could provide qualitative indicators for this construct measure [17]. An unstructured, qualitative interview, with guiding questions about these officials' political will to enact these policies would allow him to interpret their emotions and experiences [19]. Interviewing these actors could elicit subjective knowledge about their political will to implement IZ policies, which would allow Kontokosta to form a judgement about this subjectivity [19]. This subjectivity could add qualitative indicators to the conceptualization of the political will to enforce IZ policies, and these indicators contribute to this concept as a measure.
Walks and Brian's and Kontokosta's ambiguity with construct measures should not surprize political scientists, as ambiguity, overall, is a common flaw in case studies. Works in political science need to "carefully define" the key terms they employ [15], but case studies engage so closely with their contexts that scholars often deny the relevance of defining concepts. Their alibi is that they wish to keep with their scene or narrative, for a narrative explains a sequence of steps or stages, whereas definitions disrupt the narrative's flow [16] 21 . These scholars also argue that definitions fail to explain action and capture processes: much easier to describe an event than define all of its features.
Despite this argument, the shortage of definition is problematic because concepts have multiple meanings. The problem worsens in policy studies, which depend on multidisciplinary research [19]. Policy analysts often borrow terms from "other language communities" [15]. For instance, Walks and Brian [4] discuss securitization throughout their work from a political economics perspective. Nevertheless, securitization is also an international security term, and it relates to political economy when analysts or officials classify an economic issue as a security issue [20]. While Walks and Brian briefly define securitization (2015, 1624), they do not differentiate their use of it from the international security term. Without a successful rendition, their argument becomes ambiguous.
A successful rendition requires that researchers define, explain, categorize, and differentiate their core concepts, both with reference to researchers' objectives and the scholarly context within which they argue. Scholars can render their core concepts as suits their research purpose, but they must also consider their audience's needs and familiarity; otherwise, their arguments become unintelligible [15,16,21] 22 . Hence, Adcock and Collier argue that core concepts need a fleshed-out account, which "requires not just a one-sentence definition, but a broader specification of [their] meaning and entailments" (2001,532). What Adcock and Collier mean by broader specification is the above four criteria for successful renditions. This expository method is necessary to make arguments familiar, and, thus, intelligible.
The absence of differentiation is just one example from Walks and Brian [4], but the root problem for both articles is a deficiency of exposition. At the foundation of exposition is familiarity, which dictates the need to define terms when they are uncommon to an audience [15]. In Kontokosta [3], Walks and Brian [4], they used an excess of technical terms from various disciplines: political economy, geography, and law, to name a few [3,4]. Since these authors are analyzing policy, they speak to no one audience. Thus, no audience would be familiar with all these technical terms.
To strengthen their arguments, they must limit the number of unfamiliar terms and define the ones in use. Merely defining all the technical terms would detract from the case study's purpose of explaining events [15]. Also, the articles would become lengthy to satisfy the demands of all these definitions. Richard Robinson confirms this strategy; he argues that "[t]he supreme rule of stipulation … is surely to stipulate as little as possible" [15]. For these reasons, Kontokosta [3], Walks and Brian [4] must refrain from using excessive terms that need definition. This strategy frees space for the renditions they need without digressing in their case studies.
The research question or problem in both articles was not expository, so the authors can only admit as much definition as suits their purpose. Kontokosta's research problem concerns the lack of data to evaluate IZ policies [3]. This article's core concepts include geocoding, IZ policies, and spatial distribution [3]. There are also crucial programs and agencies that require definition and detail to execute Kontokosta's research design [3] 23 . He cannot cut these terms because they are integral to his study, but he needs to conceptualize them. Hence, to create room for these additions, he has to reduce the number of propositions from four to two.
Of the four propositions, Kontokosta must excise propositions three and four. These hypotheses did not coordinate with his overall research design. Proposition three predicts that a "higher proportion of IZ units will be built in areas where more multi-family housing is built" [3]. Nonetheless, the research objective was to examine the differences of IZ policies in Montgomery County with those in Suffolk County [3]. While he clarifies in his introduction that IZ policy outcomes vary due to "external factors," he is unclear why the building of multi-family housing is a relevant external factor (2015, 570). Additionally, the relationship between IZ units and multi-family housing plays no part in his discussion of results and policy implications (2015, 585-588). This proposition was unnecessary.
With proposition four, Kontokosta conveys more content than his research design can support. Proposition four states that "a regional IZ program is more likely to be coordinated with regional Smart Growth policies than local IZ programs" [3]. Smart Growth policies require 22 As in the case of Walks and Clifford [4], securitization needs differentiation because of its multiple meanings in political science. 23 One is the Moderately Priced Dwelling Unit Program. developers to build new residences in areas with existing infrastructure [22]. By reusing old infrastructure, Smart Growth policies increase sustainability. He admitted this discussion to comprehend which variables impacted the disparity in IZ unit construction (2015, 586). Seeing that Montgomery's Smart Growth policies are a potential explanation for its higher levels of IZ units, this admission is understandable. Nevertheless, with this proposition, he sacrifices depth for breadth. He even leaves Smart Growth policies undefined when they require a rendition to integrate them into his argument.
Walks and Brian's case is no better. Their research objective is to explain the processes of financialization and neoliberalization [4] 24 . Neoliberalization changes a state that is active in the market to one that is only minimally active [23]. Since Walks and Brian do not define this concept, we inferred this definition from an external source and a careful reading of Walks and Brian [4] 25 . Without inference, neoliberalization is unknowable. Walks and Brian show how the Canadian government went from implementing welfare housing policies to only regulating the housing market (2015, 1631). They show the process of neoliberalization, but they fail to explain or define it.
A contradiction arises from Walks and Brian's ambiguity. Neoliberalism involves minimal state-involvement (2015, 1624), but then they go on to argue that neoliberalization is "state-centred" (2015, 1625). To clarify this contradiction, they need to define and distinguish neoliberalization. While neoliberalization prefers financial controls over direct involvement in the housing market, it is state-centred when it over-regulates with financial controls 26 . They prove this point through their case study, but they do not differentiate minimal stateinvolvement in neoliberalism from the state-centred neoliberalization (2015, 1624, 1639). Their research question involves neoliberalization, but the conceptualization is missing.
To render this concept, Walks and Brian need to free up space because merely adding more definition would detract from their case study. To do so, they must cut as many technical terms as possible. Their article already has an excess of indefinite concepts, such as reregulation and collateralized debt obligations (2015, 1624). Their argument depends on reregulation, so they can keep this concept so long as they render it 27 . On the other hand, collateralized debt obligations they merely allude to, and it is insignificant for their argument. They must cut unnecessary terms like collateralized debt obligations and conceptualize those they depend on.
The excess of technical terms and lack of exposition runs throughout both Kontokosta [3] and Walks and Brian [4]. Nevertheless, our criticism is less towards them and more towards the case study as a method. Yin comments that many case studies become excessively long narratives. Though he omits discussion of their inadequate conceptualization, if a case study is a long, unobstructed narrative, this must mean that exposition is scarce (2003, 10). Nevertheless, due to these lengthy narratives, he argues that case studies often "result 24 They define financialization as the favouring of financial controls in a capitalist economy. 25 See David Kotz for a discussion on neoliberalism. While Walks and Clifford do not define neoliberalization, they outline neoliberalism's key tenets: "private property, commodification, and trust in price signals to provide valid information regarding underlying values, needs, and preferences, combined with antagonism toward the welfare state and redistributive policies" [23]. 26 Mortgage securitization is one control. 27 They argue that the Canadian government exacerbated the housing crisis by repeatedly implementing mortgage securitization, despite its initial failures. This repetition is what they mean by reregulation (2015, 1627, 1638 in massive, unreadable documents" (2003,10). He continues, "[t]his complaint may be appropriate, given the way case studies have been done in the past …, but this is not necessarily the way case studies must be done in the future" (2003, 10). Yin argues that case studies do not need the traditional, long narrative format, but they can integrate expository methods (1994, 134) 28 . We acknowledge Yin's criticism, and we add that these studies become intelligible with a reduction in unfamiliar terms and the conceptualization of core concepts.

Conclusion
Again, no research method is sufficient on its own; instead, political scientists need an all-round methodology. Balancing their modes provides just that. Kontokosta [3], Walks and Brian [4] illuminate this argument. The former employs a comparative case study, the latter a descriptive case study, but both are missing conceptualization. We justified our criticism due to this gap. Both articles offer new insights into housing policy, as they deeply grasp the contexts they studied, but they traded clarity for depth and specificity [3,4]. This sacrifice casts doubt on their measurement validity, for the limitations of what they seek to measure are indefinite. To provide these parameters, they needed to incorporate interviews and conceptualization, which would have gone hand-in-hand. Readers cannot determine if these authors accurately measure their content with their methods. Without conceptualization, it is also uncertain if Kontokosta and Walks and Brian know if their modes actually measure their content. Exposition and interviews were the answer.
Moreover, both articles tried to carry more technical terms than they could hold. They need to drop the excess in terms and balance their case studies with exposition. In this way, Kontokosta and Walks and Brian could familiarize their audience with their terminology and advance their arguments. Kontokosta must excise propositions three and four. Proposition three was unnecessary for his research objectives, and proposition four was too forced (2015, 572-574). Measurement validation ranks higher than forcing these hypotheses into his argument. Walks and Brian had several terms they could cut; the concept that stood out was collateralized debt obligations, which they refer to in passing (2015, 1624). Revisions like this make for better choices. For instance, Walks and Brian needed to differentiate what they mean by neoliberalization (2015, 1625). These adjustments would open the opportunity to render their core concepts successfully. Nevertheless, these changes coordinate with both articles' research design, and complementing the expository methods with their case studies would not change their findings significantly. If Kontokosta and Walks and Brian made these critical choices, then their overall inquiries become more strategic.
The housing crisis is serious, so the effectiveness of the methodology to study it should correspond. For readers to understand the failures in affording housing initiatives or housing policy corruption, scholars need to successfully conceptualize their terminology [3,4]. Without expository methods, case studies like Kontokosta's and Walks and Brian's degenerate into unclear documents [13]. Possibly, unbalanced methods are over-criticized, but if the criticism is valid, and researchers continue to ignore it, then political scientists are justified in their criticism [24][25][26][27]. Our judgement on Kontokosta [3], Walks and Brian [4] contributes to this line of thought.