diff --git a/icu4c/source/i18n/csrmbcs.cpp b/icu4c/source/i18n/csrmbcs.cpp
index 4c5bdfa560b..b9f268ab3f4 100644
--- a/icu4c/source/i18n/csrmbcs.cpp
+++ b/icu4c/source/i18n/csrmbcs.cpp
@@ -146,7 +146,6 @@ CharsetRecog_mbcs::~CharsetRecog_mbcs()
 }
 
 int32_t CharsetRecog_mbcs::match_mbcs(InputText *det, const uint16_t commonChars[], int32_t commonCharsLen) const {
-    int32_t singleByteCharCount = 0;
     int32_t doubleByteCharCount = 0;
     int32_t commonCharCount     = 0;
     int32_t badCharCount        = 0;
@@ -160,9 +159,7 @@ int32_t CharsetRecog_mbcs::match_mbcs(InputText *det, const uint16_t commonChars
         if (iter.error) {
             badCharCount++;
         } else {
-            if (iter.charValue <= 0xFF) {
-                singleByteCharCount++;
-            } else {
+            if (iter.charValue > 0xFF) {
                 doubleByteCharCount++;
 
                 if (commonChars != 0) {
diff --git a/icu4c/source/i18n/nfsubs.cpp b/icu4c/source/i18n/nfsubs.cpp
index 70ac18be468..7e50a81ca26 100644
--- a/icu4c/source/i18n/nfsubs.cpp
+++ b/icu4c/source/i18n/nfsubs.cpp
@@ -1310,10 +1310,8 @@ NumeratorSubstitution::doParse(const UnicodeString& text,
         // compute the 'effective' base and prescale the value down
         int64_t n = result.getLong(status); // force conversion!
         int64_t d = 1;
-        int32_t pow = 0;
         while (d <= n) {
             d *= 10;
-            ++pow;
         }
         // now add the zeros
         while (zeroCount > 0) {
diff --git a/icu4c/source/i18n/transreg.cpp b/icu4c/source/i18n/transreg.cpp
index 726ad56f0f6..2d371f27c37 100644
--- a/icu4c/source/i18n/transreg.cpp
+++ b/icu4c/source/i18n/transreg.cpp
@@ -141,16 +141,10 @@ Transliterator* TransliteratorAlias::create(UParseError& pe,
             // to see whether there really are ID blocks at the beginning and end (by looking for U+FFFF, which
             // marks the position where an anonymous transliterator goes) and adjust accordingly
             int32_t anonymousRBTs = transes->size();
-            int32_t transCount = anonymousRBTs * 2 + 1;
-            if (!aliasesOrRules.isEmpty() && aliasesOrRules[0] == (UChar)(0xffff))
-                --transCount;
-            if (aliasesOrRules.length() >= 2 && aliasesOrRules[aliasesOrRules.length() - 1] == (UChar)(0xffff))
-                --transCount;
             UnicodeString noIDBlock((UChar)(0xffff));
             noIDBlock += ((UChar)(0xffff));
             int32_t pos = aliasesOrRules.indexOf(noIDBlock);
             while (pos >= 0) {
-                --transCount;
                 pos = aliasesOrRules.indexOf(noIDBlock, pos + 1);
             }
 
diff --git a/icu4c/source/i18n/unesctrn.cpp b/icu4c/source/i18n/unesctrn.cpp
index 0636fe1d149..d57e9c84708 100644
--- a/icu4c/source/i18n/unesctrn.cpp
+++ b/icu4c/source/i18n/unesctrn.cpp
@@ -175,13 +175,13 @@ void UnescapeTransliterator::handleTransliterate(Replaceable& text, UTransPositi
                                                  UBool isIncremental) const {
     int32_t start = pos.start;
     int32_t limit = pos.limit;
-    int32_t i, j, ipat;
+    int32_t i, ipat;
 
     while (start < limit) {
         // Loop over the forms in spec[].  Exit this loop when we
         // match one of the specs.  Exit the outer loop if a
         // partial match is detected and isIncremental is true.
-        for (j=0, ipat=0; spec[ipat] != END; ++j) {
+        for (ipat=0; spec[ipat] != END;) {
 
             // Read the header
             int32_t prefixLen = spec[ipat++];