Political Ideology of Disaster Reduction: Does Innovation Matter?

doi:10.21203/rs.3.rs-2483857/v1

Download PDF

Research Article

Political Ideology of Disaster Reduction: Does Innovation Matter?

https://doi.org/10.21203/rs.3.rs-2483857/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The aim of this article is to study the effect of government ideology on natural disaster prevention in developing countries, taking into account the role of innovation. The study covers a sample of 91 developing countries over the period 1990 to 2020. The methodology used refers to binary choice models, since our dependent variable takes the value 1 if the country has experienced a natural disaster at a given date and 0 otherwise. We implement here a fixed effects logit model. Our results show that for all three declinations of government ideology (Right, Left and Centre) innovation tends to increase the probability of natural disasters in developing countries. This result is consistent with the alternative measure of innovation (total factor productivity) only for left-wing governments. We can suggest that the focus should be on policies to increase building standards and encourage sound innovation.

Jel Codes : P26, Q54, Q55.

Government ideology

innovation

natural disaster

Since the seminal work of Kahn (2005), the economics of natural disasters has continued to develop, opening up a vast field of analysis. Several decades later, debates persist on the nature, causes and proliferating consequences of natural disasters, particularly in developing countries. How to control or prevent natural disasters therefore becomes a crucial issue. With this in mind, the aim of this paper is to investigate the effect of government ideology on natural disaster prevention in developing countries, taking into account the role of innovation. There are three main reasons for this study: first, the trend or magnitude of natural disasters and government ideology in developing countries. Second, the lack of work linking the two concepts. And finally, the inclusion of innovation in the relationship between the above concepts.

First of all, much work has been done on the study of the risk of natural disasters. It shows that natural disasters can be made less damaging if we are able to understand how they occur or prevent them from happening in the first place (Tselios and Tompkins, 2019, 2020). However, two observations remain. Firstly, despite the increase in natural disasters, the loss of human life remains minimal on a global scale. Secondly, future damage has serious consequences for economic development, particularly in fragile economies and those located in areas at risk (Kahn, 2005; Kellenberg and Mobarak, 2008; Flores, 2018). This refers to developing economies. Even if very low probabilities are observed in certain developing regions, the fact remains that high probabilities persist in Latin America and Asia as shown in Figure 1.

Second, several studies have focused on the role of institutions in natural disaster prevention (Kahn, 2005; Keefer et al., 2011; Persson and Povitkina, 2017). Most conclusions are unanimous on the fact that good institutions improve disaster prevention. However, very few studies have addressed the question of the role of political ideology in the prevention of natural disasters. For example, the study on the political determinants of natural disaster prevention concluded that the political ideology of the government has no significant influence on natural disaster prevention. On the other hand, Repetto et al. (2022) studies that have observed the relationship between political ideology and the occurrence of earthquakes show that political ideology is a relevant factor in predicting emergency preparedness behaviour and that it is significantly related to trust in government authorities.

Third, in the literature the role of innovation has not yet been addressed in the relationship between political ideology and natural disaster prevention. Yet, Bergek et al. (2015) demonstrate that the interaction between the political context and the innovation system can be seen as an example of structural coupling that is necessary for the provision of specific resources for the maturation of societies. In the same vein on the empirical level, they show that the political ideology of the government plays a major role in technological innovation. This according to two aspects, firstly left-wing governments generally have a weak performance in terms of innovation. Secondly, the effect of political ideology is a function of the country’s economic performance. Subsequently, some studies have demonstrated the role of innovation in mitigating the risk of natural disasters (Kaur and Sood, 2019; Benali and Feki, 2020; Sood and Rawat, 2021).

This study contributes to the analysis of the political economy of natural disasters in several ways. First, to our knowledge, it innovates in the literature between political ideology and natural disaster reduction by integrating innovation as a transition channel. Second, this study proposes to empirically test the role of innovation in the relationship between political ideology and natural disaster reduction, using two measures drawn from the work of Ma (2017) and Wang et al. (2019).

The rest of the paper is organised around six points. Section 2 presents the theoretical contours of the relationship between political ideology and natural disaster reduction. Section 3 provides a sequential literature review between the three concepts. Section 4 focuses on the methodology, while Section 5 presents the data and sample for the study. Section 6 presents the results of the study and the final section is a conclusion.

Natural disaster mortality can be influenced by government policies. They can affect how individuals perceive risk-reduction measures (Neumayer and Plümper, 2007; Plümper and Neumayer, 2009). Persons who own buildings and large structures are expected to take precautions to ensure their buildings' safety. In the event of a natural disaster, this is not always the case. Following the passage of the commissions of inquiry, some building failures are observed (Osaki and Minowa, 2001).

Three potential market failures may require government intervention. The first refers to the verification phases of prevention measures. The most common one concerns the inherent characteristics of natural disasters. These are costly to verify once prevention measures are in place. Between infrastructure owners and customers, there is an information asymmetry. In the context that only the owners are conscious of whether their projects meet the government's construction standards. Promoters also put forward quality contracts in this type of market to overcome information asymmetry. When the government opposes this approach, these markets resort to a reputational strategy. However, if small-scale failures occur during natural disasters, this strategy is compromised.

The second failure relates to the quality of the infrastructure. Government intervention may also be necessary even if the quality of construction is observable. of construction is observable. Design requirements according to established standards for incorporating preventive measures in buildings are specialised and may not be well understood by purchasers, or by the building constructors themselves. In this case, information on how to design buildings is a public good that the market may not provide sufficiently. Governments that respond to the needs of citizens can improve knowledge of the principles of designing such buildings. Government-sponsored building codes are one way to do this.

The third distortion is related to people's reactions to low-amplitude but high-probability-of-death events. Individuals are unable to choose housing that meets construction standards when they are aware of the low magnitude of natural disasters that occur in the area, according to (Kahneman et al., 1982; Keefer et al., 2011). These arguments predict that mortality from natural disasters is lower when governments have greater incentives to provide public goods. These can take the form of construction contracts and building codes, making it easier to provide quality construction in private markets. It may suggest that corrupt payments are the main reason why building codes are not enforced.

However, if this is true, high mortality countries may be overrun with quality controllers who do not do their job in exchange for bribes. This does not appear to be the case. On the contrary, there are suggestive cases that indicate that such monitors are few in number, reflecting a political disinterest in the provision of public goods (Keefer et al., 2011).

3.1 Political ideology of government and innovation

Empirically, studies on the link between government ideology and innovation focus on two issues. The first is to show how government ideology affects the innovation capabilities of countries. The second examines the different aspects that government ideology can affect in order to motivate innovation.

Focusing on the factors that affect innovation, the studies focus on showing that the policy environment plays an important role in the field of innovation and technology transfer (Flanagan et al., 2011; Grimaldi et al., 2011). The different orientations between the governing parties (whether left or right) indicate that the policy environment determined by the government is not exactly the same when the political ideology changes. Indeed, the two main trends in a government’s political ideology have very different courses of action. Right-wing governments are more supportive of innovation, because it improves the competitiveness of the economy, whileleft-wing governments are more reluctant, because innovation affects the workforce[1].

To observe whether these two broad orientations are in line, Ma (2017) tests the relationship between political ideology and the government’s capacity for innovation. The author captures the government’s innovative capacity by the number of innovations generated by sector, in the same way that the private sector uses the number of inventions and patents as indicators of research and development and innovation. These data cover a sample of 50 US states over the period 1986 to 2013. The estimation technique used is a mixed Ordinary Least Squares fixed effects and random-effects approach. The objective of this dual approach is to observe inter and intra-state effects. The results suggest that right-wing governments contribute significantly to governmental innovation, and this effect is more observed between states. These results imply that more liberal states are more innovation-oriented.

In the same vein, the study by Wang et al. (2019) establishes the implications of the relationship between government ideology and innovation for a sample of 101 countries around the world. The authors start from the observation that technological innovation is strongly linked to politics, and that economic policy is influenced by the political ideology of the ruling party. Thus, they postulate, first, that the ideology of the government plays a major role in technological innovation, as a left-wing party in general is more likely to have a negative impact on innovation performance, while a right-wing party in power favours technical progress. Second, the effect of government ideology on innovation varies with economic performance. For example, if countries have better economic performance or belong to the classification of developed countries, then the negative effect of a left-wing ruling party on innovation is more pronounced. Using the Generalized Moment Method (GMM), they estimate a dynamic model in which the measures of innovation are: the number of patents and the number of trademark applications filed. These two indicators are the most commonly used in the innovation literature. The results of this study show that it is penalising for a government to opt for a left-wing ideology. The latter negatively affects innovation in these countries. Moreover, the results are consistent with the hypothesis that in developed countries if the government in place is left-wing, innovation is weakly encouraged.

More recently, the study by Aguilera et al. (2021) examines the relationship between government ideology and innovation in state-owned enterprises for the case of 131 countries. Three working hypotheses are listed. The first one postulates that within state-owned firms, government ideology contributes to increased innovation in these firms if the government is right-wing. Secondly, the encouraging effect of government ideology on innovation in state-owned enterprises is more pronounced if the government is freer in its actions and in the pursuit of itspolicy[2]. Finally, third, the effect of government ideology on state enterprise innovation weakens as political constraints are removed[3] increase. Using a random effects model, the authors validate the above-mentioned postulates 1 and 3. This shows with regard to postulate 2 that a strong ability of the state to implement its policies can lead to conflicting behaviour within the firm.

3.2 Innovation and natural disasters

The rate of technological progress in the search for solutions to environmental problems has increased considerably in recent decades. This implies a convergence with interdisciplinary fields. One of the most pressing problems is that of natural disasters. In the latter aspect, emerging innovation capacities have increased the effectiveness of the different phases of classical disaster management strategies, especially prevention. In this line, studies linking innovation and prevention against natural disasters do not use a parametric approach.

The work of Kaur and Sood (2019) takes into account this dimension of the literature, and proposes to analyse the studies on prevention by taking into account the role of innovation. The common point of the different studies that the authors identify is the consideration of Information and Communication Technologies (ICT) as a means of prevention against natural disasters. In reviewing the literature on the subject between 2011 and 2018, they use a scientometric approach. This approach consists of evaluating the strategies and prospects of a scientific field. Several elements will motivate the authors’ choice. These elements concern publications and their annual evolution during the observation period; the contribution in the different fields; the contribution of works according to the different areas (countries or regions) and finally the productivity of institutions. They lead to the conclusion that the dissemination of information and communication technologies is attractive to researchers and governments wishing to adopt them for natural disaster management.

It is in this same vein that the study of Sood and Rawat (2021) fits in. The period of census of the works between information and communication technology and natural disaster prevention goes from 2005 to 2020. The growth of publications on the subject and the number of citations clearly show the importance of taking into account innovation (ICT) in natural disaster prevention. The conclusions of this analysis go in the same direction as that of Kaur and Sood (2019). It is necessary to promote innovation in ICTs so that governments can be more responsive to the occurrence of natural disasters.

Among the works that have used econometric approaches are those of Benali and Feki (2020). They examine the relationship between natural disasters, education, information and communication technologies (ICTs) and economic growth in a sample of 20 developed and developing countries. The study period from 1990 to 2017 allows the authors to use a staggered lag autoregressive model (SLAM). The authors show that there is a unidirectional relationship between ICT and natural disasters. The policy implication of this result is that ICTs are an integral part of the natural disaster preparedness, response and recovery process. ICTs are an effective tool in the disaster management process.

[1] These machines are able to perform tasks in the factory that are specific to workers. This process increases the likelihood that the machines will replace the existing workforce, and leads to increased unemployment. This is contrary to the socialist approach of the left-wing governments (Wang et al., 2019).

[2] High capacity states are able to gain the trust of their citizens, provide public services such as health care and education, and protect their borders and territory. (Fukuyama, 2013)

[3] Political constraints refer to another set of political institutions that affect the political discretion of political actors. (Clegg et al., 2018)

The probability of a natural disaster occurring in country i in a given year t is estimated using a conditional fixed effects model developed by Chamberlain (1980). The basic empirical model can be given as follows, based on the work of Tselios and Tompkins (2020):

$${Y}_{it}^{*}={\beta }_{0}+{\beta }_{1}{Gov}_{it}+{\beta }_{2}{Inov}_{it}+{\beta }_{3}\left({Gov}_{it}*{Inov}_{it}\right)+\theta .{X}_{it}+{\epsilon }_{it} \left(1\right)$$

Where the dependent variable ${Y}_{it}^{*}$ is a latent variable. It is a continuous unobservable variable representative of the phenomenon under study corresponding to the observability criteria of a binary result. It takes the value 1 if the country i at period t experiences a natural disaster and 0 otherwise. According to Tselios and Tompkins (2020), the fact of having a probability of occurrence of a natural disaster close to 0 refers to a good consideration of prevention against these phenomena. In the remainder of Eq. 1, the variables ${Gov}_{it}$ and ${Inov}_{it}$ represent the measures of government ideology and innovation respectively. Finally, the term X refers to a set of variables assimilated to the determinants of natural disasters. Among these determinants, we find, on the one hand, the socio-economic factors. These include the standard of living or development, and inequalities. And on the other hand, institutional factors.

According to Amemiya (1981), the results of probit and logit models are generally similar in terms of probability or in terms of coefficient estimation if one takes into account normalisation problems. However, two main demarcations in the choice of our for this study should be noted. First, the logistic law tends to assign a higher probability to events than the normal law. Secondly, the logistic law favours an easier interpretation of the parameters associated with the explanatory variables, since the coefficients resulting from the estimation of a logit or probit model are not directly interpretable in economic terms. Only the signs in these coefficients provide information on the effect of waiting on the target variable.

The study covers a sample of 91 developing countries over the period 1990 to 2020. In order to present our different indicators, we refer to the Eq. 1 presented in the previous subsection. The dependent variable (${Y}_{it}^{*}$) refers to the occurrence of natural disasters. In accordance with Tselios and Tompkins (2020). This variable takes the value (1) if the country (i) has experienced a natural disaster at some time (t). It corresponds to our prevention measure against natural disasters. According to Holzmann and Jørgensen (2001), disaster prevention is the process of reducing the likelihood of natural disasters occurring. We start from the hypothesis that all the countries undergo at a given time natural hazards, this corresponds to the value (0) of our dependent variable. However, when these hazards turn into natural disasters, causing loss of life and property, the variable takes the value (1). These data on natural disasters are taken from the Centre for Epidemic and Disaster Research (CRED).

To explain the level of disaster prevention, we use political factors as the main determinants. Government ideology is a rather difficult concept in developing economies. It refers to two main political orientations: socialism, which refers to the left, and liberalism, which refers to the right. A simple way used in the various studies to capture government ideology is to use dummy variables (Potrafke, 2017). In line with Bjørnskov (2005), we divide the government ideology measure into three modalities (Right-Left-Centre). However, in order to avoid problems of multicollinearity between these variables during estimation, we do not include them in the same model. The data on government ideology come mainly from Data of Politicals Institutions (DPI) from Cruz et al. (2021).

Since our analysis focuses on two aspects, namely the verification of direct effects on the one hand and an analysis of mitigation mechanisms on the other. For this, we use innovation measures. Two indicators for measuring innovation are chosen for this study: patents and trademark applications. According to Wang et al. (2019) the various measures of government ideology can affect innovation, while innovation can serve as an improvement in natural disaster prevention. As for patents, they represent a new and useful invention for society. Therefore, their number provides an overview of the areas where they are most likely to have an impact. However, the majority of the patents identified concern engineering. In the field of natural disasters, they are most used as seismic wave absorption systems, building control systems and landslide alarms. As far as trademark applications are concerned, they are very close to the commercial field, notably manufacturing. However, they also encompass other sectors such as services that patents are not able to cover (Wang et al., 2019). In the work the total number of applications is used as a measure of innovation. These two indicators are taken from the World Development Indicators database.

In the following, we focus on the socio-economic determinants of disaster prevention. Among these determinants is the Gross Domestic Product per capita (GDP). This is an indicator of economic development (Raschky and Schwindt, 2016; Tselios and Tompkins, 2020). It measures the level of wealth created in a country at a given time. In the same model, economic growth is included as an explanatory factor for disaster prevention. While GDP measures the level of wealth in a country, its growth rate refers to the level of development held by the inhabitants within the country.

A large population clustered at the same location can force decision-makers to improve disaster prevention. Population density is considered to be a measure of economic agglomeration (Yamamura, 2013; Tselios and Tompkins, 2020). Population density is defined as the number of persons per square mile. It is a natural risk determinant because a focused population in one site is a vulnerability factor.The unemployment rate is the proportion of the active population that is unemployed but available and looking for work. The different socio-economic indicators are taken from World Bank (2020).

To conclude with the explanatory factors of prevention, we include a factorial endowment indicator, namely human capital. This is an index of human capital, based on years of schooling and returns to education. The question is whether the level of education of the population improves prevention against natural disasters. These data are taken from the Penn World Table version 10 (Feenstra et al., 2015; Zeileis, 2021). The different properties of the variables selected for our study can be found in the annexed tables.

Table 1

Summary statistics

Variables	Observations	Means Standard deviation		Min	Max
Occurrence	3974	.474 1.321 Government ideology		0	18
Right	4278	.149 .356		0	1
Left	4278	.244 .43		0	1
Center	4278	.092 .289		0	1
Without orientation	4278	.472 .499 Socio-economic indicators		0	1
GDP	3928	5923,397	9495,837	162,872	66072,842
Density	4192	173,701	589,755	1,406	8044,526
Human capital	3090	2,119	0,587	1,03	4,352
Unemployment	3960	7,375	5,842 Innovation	.11	37,97
Patents	1707	8136,342	74775,481	1	1393815
Applications for registration	2239	21422,281	91125,06	1	2104414

Source: authors

The results presented in this section are intended to give credibility to the various analyses made in the previous sections. To this end, an analysis of the basic results is presented in the first section. In the second, we test the robustness of our results.

6.1 Baseline results

To analyse the effects of government ideology on disaster prevention, we used a logit model. The application of this model requires that the dependent variable is binary. In accordance with the work of Raschky and Schwindt (2016); Tselios and Tompkins (2020) we take thevalue 1 if the country (i) has known at time (t) a natural disaster and 0 otherwise. A difficulty encountered in the implementation of binary choice models is that the coefficients derived fromthe estimates. It is recommended that marginal effects, relative risk ratios4 or even chance ratios[4] (Verbeek, 2008). In our case, it is preferable to use the odds ratios since the sample tends to be large.

Taking into account the innovative factors amounts to verifying the mediation of these factors in the relationship between government ideology and natural disaster prevention. In the table, the measure of innovation used is the number of patents. The different models in the table 2 all have an interaction term. It results from the intersection between the modality of the government’s ideology and the chosen measure of innovation. In this analysis, we focus on the coefficient of the variable (Interaction). It tells us about the sensitivity of the relationship between government ideology and natural disaster prevention.

In the tables 2 and 3 we find that the odds ratios of Right (column 1 tables 2 and 3) and Center (column 3 table 2) governments in improving disaster prevention are greater than unity. This result can be justified by the fact that governments have greater incentives to provide public goods. These can take the form of These can take the form of construction contracts and building codes, facilitating the provision of construction quality in private markets (Keefer et al., 2011). With regard to the use of patents, our results suggest that the use of patents increases the chances of each type of government to prevent natural disasters. Indeed, the coefficients associated with the patent variable are significant at the 1% level and above unity. These results are in line with those of Miao (2017). This author explains that taking technology into account in the management of the risk of natural disasters makes it possible to considerably mitigate the effects of the latter. However, a necessary condition for achieving such an objective is a considerable level of investment in the innovation sector.

Table 2

Mediating effect of the number of patents in the relationship between ideology and natural disaster prevention

Variables		Occurrence of natural disasters
Variables	(1)	(2)	(3)	(4)
Right	2.101** (0.673)
Left		0.671 (0.165)
Center			4.234*** (2.193)
Without orientation				0.573** (0.155)
Patents	1.329***	1.214***	1.299***	1.389***
	(0.041)	(0.038)	(0.036)	(0.054)
Interaction	0.917 (0.051)
Interaction		1.284*** (0.067)
Interaction			0.890 (0.075)
Interaction				0.882** (0.044)
GDP	0.714***	0.768***	0.740***	0.725***
	(0.052)	(0.051)	(0.050)	(0.052)
GDP growth	0.995	0.991	0.994	0.994
	(0.016)	(0.016)	(0.016)	(0.015)
Density	1.086*	1.091*	1.086*	1.127**
	(0.050)	(0.052)	(0.047)	(0.062)
Human capital	0.615***	0.626***	0.567***	0.482***
	(0.077)	(0.084)	(0.083)	(0.072)
Unemployment	0.952***	0.938***	0.952***	0.946***
	(0.013)	(0.015)	(0.013)	(0.014)
Constant	12.205***	10.350***	12.294***	24.316***
	(7.587)	(5.663)	(6.986)	(14.061)
Observations	1,473	1,473	1,473	1,473
Time dummy	yes	yes	yes	yes
Log -likelihood	-1013	-1013	-1013	-1013
LR Chi2	223.2	288.6	404.5	230.6
Prob Chi2	0	0	0	0
Pseudo R-squared	0.114	0.132	0.119	0.154

Note: The numbers in brackets represent standard deviations. ***, **, * represent the significance levels of 1 %, 5 % et 10 %. Source : Authors.

Table 3

Mediating effect of applications in the relationship between ideology and natural disaster prevention

Variables		Occurrence of natural disasters
Variables	(1)	(2)	(3)	(4)
Right	2.155 (2.381)
Left		0.057*** (0.040)
Center			0.362 (0.907)
Without orientation				18.413*** (15.514)
Application	1.979***	1.769***	1.912***	2.281***
	(0.116)	(0.087)	(0.107)	(0.167)
Interaction	0.947 (0.115)
Interaction		1.407*** (0.111)
Interaction			1.217 (0.327)
Interaction				0.669*** (0.061)
GDP	0.542***	0.557***	0.551***	0.558***
	(0.041)	(0.038)	(0.033)	(0.033)
GDP Growth	0.988	0.988	0.986	0.990
	(0.012)	(0.014)	(0.013)	(0.013)
Density	1.041	1.044	1.039	1.084*
	(0.050)	(0.044)	(0.040)	(0.049)
Human capital	0.808	0.846	0.767**	0.773**
	(0.116)	(0.105)	(0.096)	(0.098)
Unemployment	0.984*	0.980**	0.986	0.984
	(0.010)	(0.009)	(0.010)	(0.010)
Constante	0.400*	0.830	0.539	0.112***
	(0.204)	(0.387)	(0.312)	(0.080)
Observations	1,851	1,851	1,851	1,851
Time dummy	yes	yes	yes	yes
Log -likelihood	-1253	-1253	-1253	-1253
LR Chi2	379.6	424.3	370.4	595.8
Prob Chi2	0	0	0	0
Pseudo R-squared	0.157	0.162	0.163	0.178

Note: The numbers in brackets represent standard deviations. ***, **, * represent the significance levels of 1 %, 5 % et 10 %. Source : Authors.

For the case of applications, the results are similar in terms of the effect on natural disaster prevention (Table 3). Indeed, the coefficients of this variable are all significant and greater than unity. Thus, the application of the application allows for an improvement in disaster prevention. Turning to the interaction variable, let us observe its coefficients. We see that this coefficient is significant only for Left-wing governments and those without an ideology in the case of patents and applications (Columns 2 and 4 of the tables 2 and 3). In other words, in a context of strong innovation, Left governments tend to be better at preventing natural disasters. In contrast, the result is the opposite for non-ideological countries. As we have shown in the correlation analysis, the reducing effect in non-ideological governments can be explained by the fact that they are mostly located in low disaster risk areas.

Table 4

Marginal effects of government ideology on prevention conditional on innovation

	(1)	(2)	(3)
Variables	Right	Left	Center
Low level of innovation for patents	0.124**	-0.059	0.275**
	(0.049)	(0.038)	(0.112)
High level of innovation for patents	-0.050	0.196***	-0.026
	(0.066)	(0.046)	(0.092)
Observations	1,473	1,473	1,473
Low demand for applications	0.003	-0.006**	-0.002
	(0.005)	(0.003)	(0.004)
High demand for applications	-0.001	0.065***	0.044**
	(0.027)	(0.020)	(0.020)
Observations	1,851	1,851	1,851

Note: The numbers in brackets represent standard deviations. ***, **, * represent the significance levels of 1 %, 5 % et 10 %. Source: Authors.

The analysis of the interaction coefficients shows us that Left governments are more likely to prevent natural disasters when innovation is taken into account. However, in the correlation analysis we found that the relationship between the concepts could be positive when the level of innovation increases. In the tables 4 containing the average effects specific to the relationship between government ideology and prevention, we observe two trends conditional on innovation. For low levels of innovation (corresponding to a very low number of patents) the effect of government ideology worsens prevention against natural disasters, especially for the Right and Center governments (columns 1 and 3). We observe that this result seems to be significant in the context of applications for Left governments (column 2 table 4). On the other hand, when the level of innovation becomes high (corresponding to a high level of patents) the effect of ideology worsens disaster prevention only for Left governments (column 2).

This analysis can also be done by looking at the conditional effects graph for each type of innovation. On the graphs 2 and 3. We can see that the evolution of the marginal effects for patents and applications tends to be insignificant when the number of patents and applications increases for the Right and Centre governments. On the other hand, we observe that the curve seems to bend slightly for the case of patents (graph 2), before tending significantly towards a probability equal to unity. This result shows that the increase in the number of patents increases the probability of occurrence of natural disasters. As a result, they worsen prevention in Left governments. For the case of applications, the effect seems to stabilise towards a probability close to 0 when these applications are less than 6 in Left governments (graph 3). This shows that there may be a threshold at which innovation is detrimental to disaster prevention in developing countries.

For the control variables in the tables 2 and 3, the results are broken down as follows. In the different models in the tables 2 and 3, GDP has odds ratios below unity and significant at the 1% level. This suggests that, according to the different variations of government ideology, income level allows people to reduce the effects of natural disasters on their well-being. This supports the idea of Kahn (2005) and Kellenberg and Mobarak (2008) that income is a significant determinant of disaster risk reduction. The agglomeration factor measured here by population density has a coefficient greater than 1 and is significant at the thresholds (10% and 1%) in the table 2 and at the 10% threshold in the table 3 in column 4. This result shows that the population in the countries in our sample is not concentrated in areas at risk. This reduces government ownership in certain areas. This justifies a negative effect of the density variable on the occurrence of natural disasters. As for the human capital index, it has a hazard ratio below unity and significant at the threshold of 1% in all the model in the table 2. While in the table 3 the coefficient is significant at the 5% level in models 3 and 4. According to Tselios and Tompkins (2020), countries with low levels of education tend to have a higher probability of experiencing natural disasters. Finally, low unemployment is associated with low poverty. This leads to the abandonment by the population of the areas at risk.

6.2 Robutsness analysis

To test the robustness of the mediating effects, we use an alternative measure of innovation. This is total factor productivity. It is an indicator of technological change and innovation (Tselios and Tompkins, 2019, 2020). It is the growth in output that cannot be explained by changes in the quantity of inputs. This productivity is generally associated with the competitiveness ofan economy. By first conducting a correlation analysis between our three concepts[5], we obtain almost similar patterns to the same correlation analysis for the basic measures of innovation. In the annexed graph 4, countries without political ideology have the highest level of total factor productivity. A plausible explanation for this correlation is that, when looking at most countries without an ideology, one realises that their income comes mainly from the exploitation of natural resources. Most of these exploitations are carried out by foreign firms that import their technology. If we look at the correlation between total factor productivity and the probability of natural disasters, we find that it is initially positive for very low values of productivity. However, this relationship changes direction when total factor productivity reaches (1). This seems to support our basic analysis of innovation.

In the table 5, the different ways of validating the estimation technique are conclusive. The probability of the likelihood ratio test is below the conventional threshold of 1%. This allows us to reject the null hypothesis of null coefficients of the explanatory variables. Moreover, the pseudo R-squared tends to be very low, hence the validation of the goodness of fit of the model. Econometrically, our results show that both Right and Left governments have significant coefficients at the conventional 1 per cent level. This means that Right-wing governments are 6.225 times more likely to prevent natural disasters than other government ideology modalities. In contrast, Left-wing governments are 0.227 times less likely to prevent natural disasters than other modalities of government ideology. These findings are similar to those obtained in the baseline analysis for mediating effects. We also observe that the coefficient on the factor productivity variable is less than unity for the Left and Center governments.

Table 5

Mediating effect of total factor productivity in the relationship between ideology and disaster prevention

Variables	(1)	(2)	(3)
Right	6.225*** (4.289)
Left		0.227*** (0.114)
Center			0.851 (0.942)
Total factor productivity	0.678	0.372***	0.523***
	(0.181)	(0.105)	(0.116)
Interaction	0.222** (0.157)
Interaction		7.978*** (4.397)
Interaction			2.615 (3.047)
GDP	0.934	0.955	0.981
	(0.053)	(0.061)	(0.057)
GDP Growth	1.028**	1.022**	1.024**
	(0.013)	(0.010)	(0.011)
Density	0.980	0.991	0.986
	(0.035)	(0.036)	(0.035)
Human capital	1.370***	1.411***	1.270**
	(0.138)	(0.163)	(0.137)
Unemployment	0.983**	0.977**	0.983**
	(0.008)	(0.009)	(0.008)
Constant	0.761	1.032	0.778
	(0.288)	(0.446)	(0.287)
Observations	2,138	2,138	2,138
Time dummy	yes	yes	yes
Log -likelihood	-1378	-1378	-1378
LR Chi12	104.2	190.9	125.1
Prob Chi2	0	0	0
Pseudo R-squared	0.0338	0.0410	0.0359

Note: The numbers in brackets represent standard deviations. ***, **, * represent the significance levels of 1 %, 5 % et 10 %. Source: Authors.

Looking closely at the interaction terms, we find that only the coefficient of the interaction term between Left-wing government and total factor productivity is significant at the 1% level and above unity. Thus, for acceptable levels of total factor productivity, Left governments are able to prevent the risk of natural disasters. This result is similar to the one obtained in the baseline analysis. This result can also be confirmed by the analysis of the marginal effects table. Indeed, in the table 6 we observe that for low levels of total factor productivity only Left governments manage to implement measures that can prevent natural disasters. While in Right-wing governments the effect is the opposite. As the level of total factor productivity rises, right-wing governments catch up in disaster prevention. Left-wing governments, on the other hand, become lax in prevention. Similar conclusions can be drawn from the graph showing the evolution of marginal effects in the appendix. There is thus a threshold at which innovation undermines disaster prevention in left-wing governments.

Table 6

Marginal effects of government ideology on prevention conditional on total factor productivity

	(1)	(2)	(3)
Variables	Right	Left	Center
Low level of total factor productivity	0.345***	-0.235***	0.007
	(0.117)	(0.079)	(0.210)
High level of total factor productivity	-0.169**	0.612***	0.416
	(0.084)	(0.127)	(0.342)
Observations	2,138	2,138	2,138

Note: The numbers in brackets represent standard deviations. ***, **, * represent the significance levels of 1 %, 5 % et 10 %. Source: Authors.

[4] It is a ratio of the probability of an event occurring in persons or traits susceptible to an event, and the probability of an event occurring in persons or traits not susceptible to an event.

[5] This involves correlating government ideology with total factor productivity and then correlating total factor productivity with the probability of occurrence of natural disasters.

The major challenge of this study was to determine the effect of government ideology on natural disaster reduction in developing countries. Based on the literature on the relationship, we proceeded in two phases. First, the establishment of a theoretical link between the two concepts and then empirical verification. With regard to the theoretical link, it should be noted that very few studies have attempted to establish it. With this in mind, we have established a mediating effect through innovation. The literature in question puts forward the idea that right-wing governments are more favourable to innovation than left-wing ones. Empirically, we use a binary choice model to estimate the probability of occurrence of natural disasters. The results show that, in a direct way, the ideology of the government significantly affects the prevention of natural disasters. Taking innovation into account in the relationship, we find that innovation has a significant and destabilizing effect on prevention measures, when it tends to increase. In view of these results, we can suggest that the focus should be on policies to increase building standards and encourage sound innovation.

Funding

The authors did not receive support from any organization for the submitted work;

Competing interests

The authors have no competing interests to declare that are relevant to the content of this article.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Author Contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by and Donald Ferdinand OKERE ATANGA. The first draft of the manuscript was written by Desire AVOM, Itchoko Motande MONDJELI MWA NDJOKOU and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Ethical Statement for Solid State Ionics

Hereby, I consciously assure that for the manuscript Government ideology and natural disaster reduction: the role of innovation the following is fulfilled:

Aguilera, R., Duran, P., Heugens, P., Sauerwald, S., Turturea, R., and VanEssen, M. (2021). State ownership, political ideology, and firm performance around the world. Journal of World Business, 56(1):101–113.
Amemiya, T. (1981). Qualitative response models: A survey. Journal of economic literature, 19(4):1483–1536.
Benali, N. and Feki, R. (2020). The relationship between natural disasters, education, ict and economic growth: empirical evidence from ardl bounds testing approach. Journal of Economic Development, 45(4):93–111.
Bergek, A., Hekkert, M., Jacobsson, S., Markard, J., Sandén, B., and Truffer, B. (2015). Technological innovation systems in contexts: Conceptualizing contextual structures and interaction dynamics. Environmental Innovation and Societal Transitions, 16:51–64.
Bjørnskov, C. (2005). Does political ideology affect economic growth? Public Choice, 123(12):133–146.
Blondet, M., Garcia, G. V., Brzev, S., Rubiños, A., et al. (2003). Earthquake-resistant construction of adobe buildings: A tutorial. EERI/IAEE world housing encyclopedia.
Clegg, L. J., Voss, H., and Tardios, J. A. (2018). The autocratic advantage: Internationalization of state-owned multinationals. Journal of World Business, 53(5):668–681.
Cruz, C., Keefer, P., and Scartascini, C. (2021). Database of political institutions 2020 (dpi2020). Inter-American Development Bank. Numbers for Development.
EM-DAT (2018). Natural disasters 2018. Centre for Research on the Epidemiology of Disasters, Louvain, Belgium.
Feenstra, R. C., Inklaar, R., and Timmer, M. P. (2015). The next generation of the penn world table. American economic review, 105(10):3150–82.
Flanagan, K., Uyarra, E., and Laranja, M. (2011). Reconceptualising the ‘policy mix’for innovation. Research policy, 40(5):702–713.
Flores, A. Q. (2018). Protecting people from natural disasters: political institutions and oceanoriginated hazards. Political Science Research and Methods, 6(1):111–134.
Fukuyama, F. (2013). What is governance? Governance, 26(3):347–368.
Grimaldi, R., Kenney, M., Siegel, D. S., and Wright, M. (2011). 30 years after bayh–dole: Reassessing academic entrepreneurship. Research policy, 40(8):1045–1057.
Holzmann, R., and Jørgensen, S. (2001). Social risk management: A new conceptual framework for social protection, and beyond. International Tax and Public Finance, 8(4), 529-556.
Kahn, M. E. (2005). The death toll from natural disasters: the role of income, geography, and institutions. Review of economics and statistics, 87(2):271–284.
Kahneman, D., Slovic, S. P., Slovic, P., and Tversky, A. (1982). Judgment under uncertainty: Heuristics and biases. Cambridge university press.
Kaur, A. and Sood, S. K. (2019). Analytical mapping of research on disaster management, types and role of ict during 2011–2018. Environmental Hazards, 18(3):266–285.
Keefer, P., Neumayer, E., and Plümper, T. (2011). Earthquake propensity and the politics of mortality prevention. World Development, 39(9):1530–1541.
Kellenberg, D. K. and Mobarak, A. M. (2008). Does rising income increase or decrease damage risk from natural disasters? Journal of urban economics, 63(3):788–802.
Ma, L. (2017). Political ideology, social capital, and government innovativeness: evidence from the us states. Public Management Review, 19(2):114–133.
Miao, Q. (2017). Technological innovation, social learning and natural hazard mitigation: evidence on earthquake fatalities. Environment and Development Economics, 22(3):249–273.
Neumayer, E. and Plümper, T. (2007). The gendered nature of natural disasters: The impact of catastrophic events on the gender gap in life expectancy, 1981–2002. Annals of the association of American Geographers, 97(3):551–566.
Osaki, Y. and Minowa, M. (2001). Factors associated with earthquake deaths in the great hanshin-awaji earthquake, 1995. American journal of epidemiology, 153(2):153–156.
Persson, T. A. and Povitkina, M. (2017). “gimme shelter”: The role of democracy and institutional quality in disaster preparedness. Political Research Quarterly, 70(4):833–847.
Potrafke, N. (2017). Partisan politics: The empirical evidence from oecd panel studies. Journal of Comparative Economics, 45(4):712–750.
Raschky, P. A. and Schwindt, M. (2016). Aid, catastrophes and the samaritan’s dilemma. Economica, 83(332):624–645.
Repetto, P., Cordón, P., and Bronfman, N. (2022). The relationship between political ideology and current earthquake and tsunami preparedness. Global Environmental Change, 72:102444.
Sood, S. K. and Rawat, K. S. (2021). A scientometric analysis of ict-assisted disaster management. Natural hazards, 106(3):2863–2881.
Tselios, V. and Tompkins, E. L. (2019). What causes nations to recover from disasters? an inquiry into the role of wealth, income inequality, and social welfare provisioning. International journal of disaster risk reduction, 33:162–180.
Tselios, V. and Tompkins, E. L. (2020). Can we prevent disasters using socioeconomic and political policy tools? International Journal of Disaster Risk Reduction, 51:101764.
Verbeek, M. (2008). A guide to modern econometrics. John Wiley & Sons.
Wang, Q.-J., Feng, G.-F., Chen, Y. E., Wen, J., and Chang, C.-P. (2019). The impacts of government ideology on innovation: What are the main implications? Research Policy, 48(5):1232–1247.
World Bank. (2020). World Development Indicators. Washington, D.C.: World Bank.
Yamamura, E. (2013). Public sector corruption and the probability of technological disasters.
Economics of Governance, 14(3):233–255.
Zeileis, A. (2021). pwt10: Penn World Table (Version 10.x). R package version 10.0-0.

No competing interests reported.

Annex.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Political Ideology of Disaster Reduction: Does Innovation Matter?

Status:

Version 1

Abstract

Figures

1. Introduction

2. Framework

3. Literature Review

4. Methodology

5. Data

6. Empiricals Results

7. Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1