What reasons make the device cost per KVAr higher for the series FACTS device (TCSC) as compared to the shunt one (SVC), since power handling capacity is the same?
This is probably due to the need of the series compensators to ride through the contingency, dynamic overloading, and (most importantly) bolted faults. These required capacities necessitates that the series connected compensators be equipped with metal oxide arresters, or solid states devices. They also have to be rated (higher) in a way that handle these events.
You can also take a look at the following paper for more information on controlled series capacitors.
Article Analysis of sub-synchronous resonance in doubly-fed inductio...
It can be shown that the ratio of the series and shunt rating can be obtained as follows:
Qse/Qsh = tan^2 (delta_max/2)
where Qse and Qsh are the ratings of the series and the shunt capacitors, respectively. Also, delta_max is the maximum angular difference between the two ends of the line. Although series capacitors tend to be twice as costly as shunt capacitors (per unit var), they are still cheaper to use, for the same purpose of power flow enhancement. Moreover, unlike the shunt capacitors, the location of the series capacitors are not critical.