20% Variable Pay in Your CTC: What Arrives Each Month

The hypothetical offer

• CTC ₹15,00,000, of which variable is 20%, a target of ₹3,00,000.
• Fixed ₹12,00,000, including ₹36,000 employer PF (12% of the ₹25,000 EPF ceiling that applies from 17 September 2026).
• Fixed gross salary is therefore ₹11,64,000, or ₹97,000 a month.
• Employee PF of ₹3,000 a month.
• The payout is assumed to fall in March 2027, inside tax year 2026-27. If your plan pays after 31 March, the payout belongs to the next tax year, whose rates are not yet known.

Monthly in-hand from the fixed part alone

Fixed gross of ₹11,64,000 less the ₹75,000 standard deduction gives ₹10,89,000 of total income. That is under ₹12,00,000, so under the new regime the section 156 rebate makes the tax nil if variable is ignored. Monthly in-hand is ₹97,000 − ₹3,000 PF = ₹94,000, plus or minus any state professional tax.

This is the figure to budget against, not ₹15,00,000 ÷ 12 = ₹1,25,000. The gap of ₹31,000 a month is made up of three things:

• ₹25,000 of variable that is not paid monthly;
• ₹3,000 of employer PF that never reaches your account;
• your own ₹3,000 PF.

What each payout level really yields

Variable pay is salary, so it adds to total income for the year it is paid in. The rates used below are from the Income Tax Department's AY 2026-27 slab page, which the Finance Bill 2026 memorandum carries unchanged into tax year 2026-27.

The 100% row works like this: ₹60,000 on the first ₹12,00,000, plus 15% of ₹1,89,000, which is ₹28,350. That gives ₹88,350 before cess and ₹91,884 after 4% cess.

The share kept is lowest on a partial payout, which surprises people. The fixed salary alone attracts no tax, so the variable carries all of the year's tax. The first ₹1,11,000 of variable is what pushes income over the ₹12,00,000 rebate line. A 70% payout sits where the section 156 marginal relief no longer caps the tax, and the full 15% slab applies.

# Synthetic ₹15 lakh CTC with 20% variable. New regime, tax year 2026-27.
BANDS = [(400000, 0), (800000, .05), (1200000, .10), (1600000, .15),
         (2000000, .20), (2400000, .25), (float("inf"), .30)]

def year_tax(total_income):
    t, lo = 0.0, 0
    for hi, r in BANDS:
        if total_income > lo:
            t += (min(total_income, hi) - lo) * r
        lo = hi
    if total_income <= 1200000:
        return 0.0
    return round(min(t, total_income - 1200000) * 1.04, 2)

FIXED_GROSS, TARGET = 1164000, 300000

def payout_view(pct):
    variable = TARGET * pct
    tax = year_tax(FIXED_GROSS + variable - 75000)
    return variable, tax, variable - tax

for pct in (0.7, 1.0, 1.2):
    v, t, net = payout_view(pct)
    print(f"{pct:.0%}: paid {v:,.0f}  tax {t:,.0f}  kept {net:,.0f}")

# Spreading the same 100% tax across the year instead of the payout month:
print("monthly TDS if projected from April:", round(payout_view(1.0)[1] / 12))

Two ways payroll can collect the same tax

Section 392(1) of the Income-tax Act, 2025 requires TDS on salary at the average rate on the employer's estimate of your salary for the tax year. How that estimate treats variable pay changes your monthly cash, but not the year's total.

The second approach smooths the tax, but it withholds ₹7,657 a month against a payout that may never come in full. If the payout is lower, March TDS falls or a refund arises when you file. Neither approach changes the final liability. Ask payroll which method they use so the March payslip does not come as a shock. The monthly mechanics are covered in more detail in how your employer calculates monthly TDS.

Questions to put to HR before signing

1. 1Which payout month, and does that month fall inside or after the tax year?
2. 2Is the variable percentage applied to CTC, to fixed pay or to basic? A "20% variable" defined on basic is a much smaller sum.
3. 3What happens if you leave before the payout date? Many plans require you to be on the rolls on the payout day.
4. 4Is the multiplier capped, for example at 120%?

A plan with individual and company multipliers needs its own reading. Reading a variable pay plan works through one. The kind of scenario modelling in the function above is ordinary work on our Data Science programme.