This study investigates consumer political ideology preferences regarding TV news programs in Taiwan’s environment of open and competitive contemporary media. We use audience ratings data for Taiwan’s news channels between 2004 and 2008 to ex-amine the sensitivity of consumers to the political ideology and accuracy of news pro-grams. The survey data and regional viewership data considered in this study suggest that consumers may prefer news channels whose slant is close to their own ideology.
However, the estimation results above may suffer from endogeneity problems re-sulting from reverse causality. Thus, we also explore changes in viewership patterns during periods with more political news to further examine consumer preferences re-garding the slant of news channels. We find that during the 2008 presidential election and the Chen Yunlin incident, the increase in ratings for TVBS was greater than that for FTV in regions with greater proportions of Pan-Blue supporters. This finding indicates that during political events, consumers are more inclined to watch news programs with a political ideology approximating their own.
Moreover, the results for other news channels are consistent with previous studies regarding the slant of each news channel. Our results show that the increase in ratings for CTI and ETT was greater than that for FTV in regions with a greater proportion of Pan-Blue supporters; in contrast, there was no significant difference between changes in ratings for FTV and SET. These findings are in accordance with the results of pre-vious studies indicating that the news content of CTI and ETT are more Pan-Blue than FTV and that both FTV and SET are Pan-Green.
After examining consumer sensitivity to political ideology, we explore the sen-sitivity of consumers to the accuracy of news programs. We employ TVBS’s Chou
Cheng-Pao videotape incident and SET’s 228 false footage incident as examples of sudden changes in news accuracy for these two news channels. We find that changes in ratings for TVBS and SET during incidents in which false reports were exposed are small or insignificant.
This paper shows that consumers tend to watch news that reflects a political ide-ology that is similar to their own in Taiwan’s current environment of open and com-petitive contemporary media. Furthermore, compared to the tendency to watch news that approximates self-beliefs, consumers are not very sensitive to accuracy in news programs.
Tables
Table 1.1
24-hour Television News Channels in Taiwan Name Abbreviation in
TVBS TVBI Company Limited (a subsidiary of Television Broadcasts Limite in Hong Kong) ERA Group of Taiwan (2004-2005)
TVBI Company Limited (2005--)
1995
FTV News
FTV Formosa Television Inc. 1997
CTi News
CTI China Times Group (2004- Nov. 2008) Want Want China Times Group (Nov.
2008--)
1997
SET News
SET Sanlih E-Television Inc. 1998
Era News ERA ERA Communications Inc. 1996
ETTV News
ETT Eastern Broadcasting Co. 1997
ETtoday ETTO Eastern Broadcasting Co. 1999
Note: ETtoday was off-air between August 7, 2005, and June 25, 2006. It changed its name to EBC Financial News on mid-December, 2008.
Table 1.2
Summary Statistics of the TEDS Survey Sample
Observations Mean S.D
Choice of TV channels:
FTV 3,584 0.182 0.386
TVBS 3,584 0.252 0.434
SET 3,584 0.112 0.315
ETT 3,584 0.099 0.299
CTI 3,584 0.096 0.295
Political ideology:
Pan-blue 3,584 0.571 0.495
Gender:
Male 3,584 0.516 0.500
Level of education:
Junior high school 3,584 0.116 0.320
High school 3,584 0.292 0.455
Junior college 3,584 0.157 0.364
Above university 3,584 0.293 0.455
Ethnic Group:
Minnanese 3,584 0.733 0.442
Mainlander 3,584 0.166 0.372
Aborigine 3,584 0.009 0.097
Notes: (1) The data samples comprise TEDS survey data from the 2004 and 2008 legislative elections, the 2005 county governor and city mayor elections, and the 2006 Taipei and Kaohsiung City mayor elections. (2) S.D. represents the standard deviation of the variable.
Table 1.3 Standard errors are in brackets. (3) The control group of “level of education” is “below elementary school.” The control group of “Ethnic Group” is “Hakka.” The control group of “year of interview” is “2004.” (4) Other control variables not presented in this table include: age, income categories, occupation, regions, and mother’s ethnicity. (5) ***Significant at 1 percent level. **Significant at 5 percent level. *Significant at 10 percent level.
Table 1.4
Choices of TV News Channels (Average Partial effects)
Choice of TV channels FTV TVBS SET ETT CTI
Notes: (1) Average partial effects are presented in this table. The reference group is FTV. (2) Standard errors are in brackets.
(3) The control group of “level of education” is “below elementary school.” The control group of “Ethnic Group” is
“Hakka.” The control group of “year of interview” is “2004.” (4) Other control variables not presented in this table include:
age, income categories, occupation, regions, and mother’s ethnicity. (5) ***Significant at 1 percent level. **Significant at 5 percent level. *Significant at 10 percent level.
Table 1.5
Consumer Political Preference in Taiwan
Proportion of Pan-Blue Proportion of Pan-Green Observations Mean Standard errors Mean Standard errors
Greater Taipei region 51 0.419 0.088 0.223 0.064
Northern region 51 0.444 0.094 0.203 0.073
Central region 51 0.379 0.072 0.219 0.056
Southern region 51 0.348 0.058 0.272 0.050
Notes: (1) The data is from monthly public opinion surveys conducted by TVBS in May, June, July, August, September, and December 2004 and from 2005 (excluding August and October) to April 2006 (excluding February), and from the public opinion survey data collected by Global Views Monthly between May 2006 and 2008. (2) We classify people who support for KMT and PFP as Pan-Blue Coalition supporters. People who support for the DPP and TSU are classified as Pan-Green Coalition supporters.
Table 1.6
Weekly Average Ratings For the News Channels
Region All Greater Taipei Northern Central Southern
Channel Mean Std. Dev. Mean Mean Mean Mean
TVBS 0.57 0.20 0.72 0.67 0.39 0.49
FTV 0.30 0.10 0.36 0.30 0.25 0.30
CTI 0.42 0.14 0.53 0.44 0.32 0.38
SET 0.44 0.14 0.57 0.41 0.36 0.42
ERA 0.21 0.06 0.26 0.22 0.17 0.19
ETT 0.34 0.12 0.44 0.38 0.26 0.29
ETTO 0.21 0.10 0.30 0.20 0.15 0.19
Notes: (1) Sample period: December 28, 2003, to December 28, 2008. (2) ETTO was off-air between August 7, 2005, and June 25, 2006. After discarding monthly ratings data that were not included in the public opinion survey data, 6,073 samples are obtained.
Table 1.7
Viewership Patterns During Political Events
Samples FTV and TVBS All news channels
Pan-blue ×TVBS 2.324*** 2.303*** 2.324*** 2.303***
08 election ×Pan-blue × CTI 3.585***
[0.910]
08 election ×Pan-blue × ERA 1.515
[0.981]
08 election ×Pan-blue × ETT 5.050***
[0.950]
08 election ×Pan-blue × SET 0.224
[1.614]
08 election ×Pan-blue × ETTO 3.743***
[1.138]
Observations 1,774 1,774 6,073 6,073
R-squared 0.857 0.859 0.861 0.863
Notes: (1) Standard errors are in brackets. (2) The dependent variable is ln( yrjt) − ln (yrot). (3) Column 1 and 2 present results using data of FTV and TVBS. Column 3 and 4 present results using data of all news channels. (4) All other controls not presented in this table include: the proportion of pan-blue supporters in a region (Brt), the square term of the proportion of pan-blue supporters in a region (Brt2), the interaction term between the exceptional event and the proportion of pan-blue supporters in a region ( Et∙ Brt ), the interaction term between the exceptional event and the square term of the proportion of pan-blue supporters ( Et∙ Brt2 ), regional fixed effects ϕr, and week-channel fixed effects kjt. (5) The control group of news channels is FTV. “Chen” represents the Chen Yun-lin incident, and “08 election” represents the 2008 presidential election. (6) ***Significant at 1 percent level. **Significant at 5 percent level. *Significant at 10 percent level.
Table 1.8 Robustness Checks I
Samples FTV and TVBS All news channels
Pan-blue ×TVBS 2.317*** 2.295*** 2.317*** 2.295***
08 election ×Pan-blue × CTI 3.595***
[0.828]
08 election ×Pan-blue × ERA 1.517*
[0.893]
08 election ×Pan-blue × ETT 5.073***
[0.865]
08 election ×Pan-blue × SET 0.225
[1.464]
08 election ×Pan-blue × ETTO 3.760***
[1.040]
Observations 1,774 1,774 6,073 6,073
R-squared 0.856 0.857 0.830 0.832
Notes: (1) In these specifications, month-channel fixed effects (the interactions between month dummies and news channel dummies) are used to control for unobservable program characteristics varying over time. (2) The dependent variable is ln( yrjt) − ln (yrot). (3) Column 1 and 2 present the result using data of FTV and TVBS. Column 3 and 4 present the result using data of all news channels. (4) Standard errors are in brackets. (5) The control group of news channels is FTV. “Chen” represents the Chen Yun-lin incident, and “08 election” represents the 2008 presidential election. (6) ***Significant at 1 percent level. **Significant at 5 percent level. *Significant at 10 percent level.
Table 1.9 Robustness Checks II
Samples FTV and TVBS All news channels
Pan-blue ×TVBS 1.220*** 1.207*** 1.220*** 1.207***
08 election ×Pan-blue × CTI 4.060***
[0.818]
08 election ×Pan-blue × ERA 2.342***
[0.868]
08 election ×Pan-blue × ETT 5.882***
[0.857]
08 election ×Pan-blue × SET 0.536
[1.384]
08 election ×Pan-blue × ETTO 4.830***
[0.989]
Observations 1,774 1,774 6,073 6,073
R-squared 0.733 0.743 0.715 0.721
Notes: (1) In these specifications, channel-specific time trends (t、t2、t3 and the interactions between these time-related variables and news channel dummies) are used to control for channel specific time trends. (2) The dependent variable is ln( yrjt) − ln (yrot).
(3) Column 1 and 2 present the result using data of FTV and TVBS. Column 3 and 4 present the result using data of all news channels.
(4) Standard errors are in brackets. (5) The control group of news channels is FTV. “Chen” represents the Chen Yun-lin incident, and
“08 election” represents the 2008 presidential election. (6) ***Significant at 1 percent level. **Significant at 5 percent level.
*Significant at 10 percent level.
Table 1.10
Robustness Checks III (Reduced Form Estimation)
Samples FTV and TVBS All news channels
Pan-blue ×TVBS 1.537*** 1.515*** 1.537*** 1.515***
08 election ×Pan-blue × CTI 3.194***
[0.322]
08 election ×Pan-blue × ERA 0.682**
[0.268]
08 election ×Pan-blue × ETT 2.965***
[0.267]
08 election ×Pan-blue × SET 0.434
[1.643]
08 election ×Pan-blue × ETTO 1.195***
[0.389]
Observations 1,774 1,774 6,073 6,073
R-squared 0.857 0.859 0.861 0.863
Notes: (1) Standard errors are in brackets. (2) The dependent variable is yrjt. (3) Column 1 and 2 present the result using data of FTV and TVBS. Column 3 and 4 present the result using data of all news channels. (4) All other controls not presented in this table include:
the proportion of pan-blue supporters in a region (Brt), the square term of the proportion of pan-blue supporters in a region (Brt2), the interaction term between the exceptional event and the proportion of pan-blue supporters in a region ( Et∙ Brt ), the interaction term between the exceptional event and the square term of the proportion of pan-blue supporters ( Et∙ Brt2 ), regional fixed effects ϕr, and week-channel fixed effects kjt(interaction terms of week dummies and news channel dummies). (5) The control group of news channels is FTV. “Chen” represents the Chen Yun-lin incident, and “08 election” represents the 2008 presidential election. (6)
***Significant at 1 percent level. **Significant at 5 percent level. *Significant at 10 percent level.
Table 1.11
Consumers' Sensitivity to News Accuracy
Samples 2007 2004 to 2008
TVBS × the week of the Chou
videotape incident was exposed -0.129 -0.076 -0.079 -0.079
[0.095] [0.086] [0.103] [0.064]
TVBS × one week after the Chou
videotape incident was exposed -0.150** -0.014 -0.096 -0.043
[0.066] [0.047] [0.075] [0.049]
TVBS × two weeks after the Chou
videotape incident was exposed -0.151 -0.153* -0.097 -0.185**
[0.108] [0.085] [0.106] [0.085]
TVBS × three weeks after the Chou
videotape incident was exposed 0.044 0.012 0.099 -0.022
[0.111] [0.090] [0.104] [0.090]
TVBS × four weeks after the Chou
videotape incident was exposed -0.057 -0.023 -0.003 -0.059
[0.050] [0.047] [0.067] [0.046]
SET × the week of the false footage
was exposed 0.213* 0.235** 0.341** 0.160
[0.120] [0.111] [0.140] [0.114]
SET × one week after the false
footage was exposed -0.130* -0.149** -0.002 -0.233***
[0.072] [0.066] [0.067] [0.060]
SET × two weeks after the false
footage was exposed -0.014 0.062 0.115 -0.031
[0.094] [0.073] [0.123] [0.078]
SET × three weeks after the false
footage was exposed -0.171** -0.123* -0.043 -0.225***
[0.076] [0.070] [0.098] [0.078]
SET ×four weeks after the false
footage was exposed -0.074 0.010 0.046 -0.101*
Notes: (1) Standard errors are in brackets. (2) The dependent variable is ln( yrjt) − ln (yrot). (3) Column 1 and 2 present the result using data of year 2007. Column 3 and 4 present the result using data of all years. (4) All other controls not presented in this table include: news channel dummies, the proportion of pan-blue supporters in a region (Brt), the square term of the proportion of pan-blue supporters in a region (Brt2), the interaction term between the exceptional event and the proportion of pan-blue supporters in a region ( Et∙ Brt ), the interaction term between the exceptional event and the square term of the proportion of pan-blue supporters ( Et∙ Brt2 ), the interaction term between the exceptional event and news channel dummy( Et∙ Ij ), regional fixed effects ϕr, week dummies in column 1and 3, t、t2、t3 and the interactions between these time-related variables and news channel dummies to control for channel specific time trends in column 2 and 4. (5) The control group of news channels is FTV. The coefficient on “TVBS/SET× the week of the Chou videotape/special reports broadcast” captures relative to FTV, ratings fluctuations for TVBS/ SET during these two false reports incidents. (6) ***Significant at 1 percent level. **Significant at 5 percent level. *Significant at 10 percent level.
Chapter 2
Long-Term Impacts of Early-life Exposure to Malaria: Evidence from Taiwan’s 1950 Eradi-cation Campaign
2.1 Introduction
Research has indicated that health conditions in early life could have long-lasting impacts on various developmental outcomes into adulthood (Case et al., 2005; Almond and Currie, 2011a). Early-life health conditions predetermine one’s health capital later in life, and health capital is a crucial element of economic competence throughout one’s life. Researchers used to believe that the placenta was an impeccable filter;
hence, it was acceptable for pregnant women to smoke or drink in the 1950s (Almond and Currie, 2011b).
Barker (1992) first formalized this conjecture by proposing the fetal origins hy-pothesis. The fetal origins hypothesis posits that ”certain chronic conditions later in life can be traced to the course of fetal development.” Consequently, certain chronic health conditions such as diabetes and cardiovascular diseases in middle or old age can be traced back to the fetal environment. Extensive evidence in the medical literature supports this hypothesis. For example, Langley-Evans (2001) and Brown et al. (2004) find that poor early fetal conditions increase the risks of vascular resistance, hyperten-sion, and schizophrenia. In addition, a series of epidemiological studies indicate that
ity and several other health problems in middle age (Roseboom et al., 2000; Roseboom et al., 2001; Ravelli et al., 2005). This evidence shows a strong relationship between the in utero environment and long-term outcomes.
However, it is not easy to demonstrate the causality between health conditions in early life and long-term outcomes. The main problem with estimating the effect of early-life health is mainly due to omitted confounders, such as socioeconomic condi-tions. To address this identification concern, researchers have established ingenious methods to mitigate potential bias from omitted confounders. For example, Almond (2006) uses the abrupt and unexpected attribute of the 1918 influenza pandemic as a natural experiment to test the fetal origins hypothesis. He compares long-term out-comes for cohorts born during the 1918 influenza epidemic to the outout-comes of cohorts born one year before or after the epidemic and finds large negative long-term conse-quences from exposure to the 1918 influenza virus, such as lower educational attament and socioeconomic status and higher disability rates. Chen and Zhou (2007) in-dicate that cohorts exposed to the China’s Great Famine in early life were significantly shorter, worked fewer hours and earned less income in adulthood. Lee (2014) also shows that prenatal exposure to the Korean War led to lower educational attainment and labor market performance and higher disability rates later in life.
In recent years, a series of studies has employed malaria exposure to estimate the effects of early-life health shocks. Some studies use the instrumental variables iden-tification strategy and employ climatic factors such as rainfall and temperature as in-struments for malaria deaths. These studies indicate that cohorts exposed to malaria in their year of birth have significantly lower levels of educational attainment, shorter height as adults, and a higher possibility of cardiovascular and lung diseases in old ages (Hong, 2007; Barreca, 2010; Chang et al., 2014). Moreover, a series of studies exploits the implementation of malaria eradication campaigns in historically endemic countries as quasi-experiments and finds that malaria exposure in early life has negative effects on one’s socioeconomic outcomes but has an ambiguous effect on education
attain-ment in adulthood (Lucas, 2010; Bleakley, 2010; Cutler et al., 2010; Barofsky et al., 2011; Venkataramani, 2012; Hong, 2013) (further discussion in Section 2.2).
Based on the aforementioned literature, this paper estimates the long-term impacts of in utero and postnatal exposure to malaria in Taiwan. In Taiwan, malaria accounted for most of the overall disease burden during the early 20th century. The death rate was approximately 0.3%. Malaria still affected people severely (especially children) in the late 1940s. In the 1950s, Taiwan implemented a nationwide malaria eradication campaign. The campaign led to a significant decline in malaria. Taiwan was certified by the WHO as an area where malaria had been eradicated in 1965, and the country has successfully maintained its malaria-free status through today.
Utilizing the malaria eradication campaign in Taiwan as a quasi-experiment, this paper estimates the long-term impacts of early-life (in utero and postnatal) exposure to malaria through the combination of geographic variation in the reduction of malaria intensity that resulted from eradication and cohort exposure based on the timing of the national malaria eradication campaign. Specifically, we match adults in the 1992 – 2012 Taiwan Social Change Survey to the malaria intensity in their individual place and year of birth in the 1950s and use the difference-in-differences identification strategy to alleviate potential biases caused by omitted factors and measurement errors
The contribution of this paper is twofold. First, in addition to examining the im-pacts of malaria exposure on own education attainment and income as discussed in existing studies, this paper investigates the effect on spousal educational attainment, which is a marriage-related outcome. To the best of our knowledge, this paper is the first that attempts to investigate the impacts of early-life malaria exposure on marriage.
Second, rather than using the pre-eradication geographic variation in malaria risk used in most previous studies on malaria eradication campaigns, this paper utilizes the ge-ographic variation in the reduction in malaria intensity between the pre-eradication period and the post-eradication period because the reduction in malaria intensity that resulted from the eradication campaign could capture the geographic variation more
precisely and allow us to directly examine the impacts of the eradication.
In addition to contributing to the research on early-life health conditions, the crucial fact that malaria continues to be a severe public health issue in developing countries today underscores the importance of this paper. Malaria occurs in nearly 100 countries today, most of which are in Africa, and more than 3 billion people are at risk of suf-fering from the disease. Children under five years of age and pregnant women are the most severely affected (WHO, 2013). This paper helps to verify the long-term impacts of malaria and evaluate malaria control policies using Taiwan’s experience during the mid-twentieth century.
Our estimates show that men born after the malaria eradication in regions with larger decreases in malaria intensity had larger increases in their own educational at-tainment, in family income, and in their spouses’ educational attainment in adulthood.
We also use the 1980 census data to show there is a sharp education increase after the eradication, and the identification strategy is very similar to the regression disconti-nuity design. Although malaria eradication increased women’s educational attainment based on data from the 1980 census, other significant benefits for women from the eradication were not observed in most specifications, and explanations for this result are discussed. Our results suggest that there were negative and significant effects of early-life malaria exposure on long-term outcomes in Taiwan.
The remainder of this paper is organized as follows. Section 2.2 describes the literature review. Section 2.3 introduces background information on malaria and the data used in this paper. Section 2.4 presents the empirical specification. Section 2.5 discusses the estimation results, and section 2.6 concludes the paper.