I have been thinking deeply about how to decide what stock to sell out of, when i want to add a new PF stock. I realized the only way is to to do the grungy work of building out some sort of a model for what returns I expect stock to give in next K years, and comparing all PF stocks based on that metric. This would require me to estimate 2 critical components for each PF stock, exit valuations, and fundamentals (earnings for non lenders and book value for lenders) growth. That, along with current valuations would lead me to estimate future CAGR and thus estimate which stock to sell. I do not like doing this exercise because there are just opinions. Well informed ones, but still opinions nonetheless. There is no way i can objectively justify why my exit multiple would be 7 for a stock and not 6. However, i still expect this exercise to be useful since I would only be using the numbers to make relative decisions between the stocks (which one to exit), not absolute ones. I still intend to apply my own judgement on top of this filtering criterion, this is not a quant procedure.
| Instrument | Percent of PF | Timewise visibility | Valuations | Est Topline Growth | Est Profit Growth | Future CAGR Expectations | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Future | ||||||||||||||
| Current | Bear | Bull | Avg | Bear | Bull | Bear | Bull | Avg | Bear | Bull | Avg | |||
| IDFCFIRSTB | 0.134 | 4 | 2.2 | 2.5 | 4 | 3.25 | 0.2 | 0.25 | 0.2 | 0.25 | 0.225 | 0.2389 | 0.4515 | 0.3505 |
| NEULANDLAB | 0.124 | 4 | 3 | 4 | 6 | 5 | 0.15 | 0.15 | 0.15 | 0.2 | 0.175 | 0.2357 | 0.427 | 0.335 |
| VAIBHAVGBL | 0.115 | 4 | 6 | 4 | 8 | 6 | 0.17 | 0.22 | 0.25 | 0.4 | 0.325 | 0.1295 | 0.5043 | 0.325 |
| RACLGEAR | 0.095 | 4 | 1.46 | 2 | 3 | 2.5 | 0.15 | 0.22 | 0.17 | 0.3 | 0.235 | 0.2657 | 0.5564 | 0.4127 |
| MASTEK | 0.089 | 4 | 2.75 | 3 | 5 | 4 | 0.15 | 0.2 | 0.17 | 0.25 | 0.21 | 0.1957 | 0.4515 | 0.3288 |
| SEQUENT | 0.068 | 4 | 5 | 4 | 7 | 5.5 | 0.15 | 0.17 | 0.2 | 0.23 | 0.215 | 0.1348 | 0.3379 | 0.2442 |
| LAURUSLABS | 0.067 | 4 | 5.4 | 4 | 7 | 5.5 | 0.2 | 0.25 | 0.25 | 0.3 | 0.275 | 0.1596 | 0.3871 | 0.2808 |
| SAREGAMA | 0.059 | 4 | 7.5 | 5 | 15 | 10 | 0.2 | 0.26 | 0.24 | 0.32 | 0.28 | 0.1204 | 0.5697 | 0.3754 |
| AXTEL | 0.048 | 4 | 3.7 | 3 | 7 | 5 | 0.15 | 0.15 | 0.2 | 0.25 | 0.225 | 0.1387 | 0.466 | 0.3207 |
| DYNPRO | 0.047 | 4 | 2.9 | 2 | 4 | 3 | 0.14 | 0.2 | 0.17 | 0.25 | 0.21 | 0.0662 | 0.3546 | 0.2202 |
| NCC | 0.045 | 4 | 0.65 | 0.4 | 1.1 | 0.75 | 0.1 | 25 | 0.1 | 0.3 | 0.2 | -0.0258 | 0.4827 | 0.2437 |
| POKARNA | 0.042 | 4 | 2.68 | 2 | 4 | 3 | 0.15 | 25 | 0.2 | 0.3 | 0.25 | 0.1153 | 0.4368 | 0.2857 |
| ANGELBRKG | 0.037 | 4 | 4 | 3 | 6 | 4.5 | 0.2 | 30 | 0.25 | 0.35 | 0.3 | 0.1632 | 0.494 | 0.3388 |
| ARMANFIN | 0.022 | 4 | 2.5 | 3 | 5 | 4 | 0.2 | 30 | 0.2 | 0.32 | 0.26 | 0.2559 | 0.5697 | 0.4171 |
| PIXTRANS | 0 | 4 | 1.8 | 2.5 | 5 | 3.75 | 0.12 | 20 | 0.15 | 0.3 | 0.225 | 0.2484 | 0.6782 | 0.4717 |
I have taken a bear and bull case for 3 parameters: exit valuations, topline growth and bottomline growth. I have estimated future CAGR based on 3 metrics: bear, bull and average where exit valuations and bottomline growth are average of bear and bull case. Based on that above table was computed. Based on my calculations, I have decided to exit NCC. Some human judgement has to be applied on top of the numbers. Even Dynemic, sequent and laurus are giving low growth. However, I am more sure of the growth in sequent and laurus than most companies so doesn’t make sense to sell those. NCC with the infra cycle revival would be most difficult to predict. As one can see from the table, I have decided to replace NCC with pix transmissions (update: Pix transmissions - low profile microcap company - #118 by sahil_vi). Investment thesis is here.
Please note that none of whatever I have posted is investment advice. This is purely to document my own thinking and share methodology with forum. Forum members must have their own buy and sell decisions based on their own individual research or suggestions of their financial advisor. All estimates are wrong. Some are useful. I hope mine prove to be useful to my portfolio.
Also attaching the excel in case someone wants to play with it.
What to Replace_.xlsx (17.7 KB)