結論

date_format() を使おう

だめな例1

library(ggplot2)
library(scales)

# データを作る
df1 <- data.frame(date = c(as.POSIXct("2016-01-01 10:00:00"),as.POSIXct("2016-01-02 6:00:00"),as.POSIXct("2016-01-02 23:00:00")),value = c(10,30,20))

# データを確認する
df1
                 date value
1 2016-01-01 10:00:00    10
2 2016-01-02 06:00:00    30
3 2016-01-02 23:00:00    20

# ggplot
ggplot(df1,aes(x=date, y = value)) + geom_point() + scale_x_datetime(breaks = date_breaks("hour"),date_labels = "%Y-%m-%d %H") + theme(axis.text.x = element_text(angle = 45, hjust = 1))

だめな例2

# データを作る
df2 <- data.frame(date = c(as.POSIXct("2016-01-01 10:00:00", tz="Japan"),as.POSIXct("2016-01-02 6:00:00", tz="Japan"),as.POSIXct("2016-01-02 23:00:00", tz="Japan")),value = c(10,30,20))
 
# データを確認する
df2
                 date value
1 2016-01-01 10:00:00    10
2 2016-01-02 06:00:00    30
3 2016-01-02 23:00:00    20 

# ggplot
ggplot(df2,aes(x=date, y = value)) + geom_point() + scale_x_datetime(breaks = date_breaks("hour"),date_labels = "%Y-%m-%d %H") + theme(axis.text.x = element_text(angle = 45, hjust = 1))

良い例1

ggplot(df1,aes(x=date, y = value)) + geom_point() + scale_x_datetime(breaks = date_breaks("hour"), labels = date_format("%Y-%m-%d %H")) + theme(axis.text.x = element_text(angle = 45, hjust = 1))

良い例2

ggplot(df2,aes(x=date, y = value)) + geom_point() + scale_x_datetime(breaks = date_breaks("hour"), labels = date_format("%Y-%m-%d %H", tz = "Asia/Tokyo")) + theme(axis.text.x = element_text(angle = 45, hjust = 1))

参考